Daily Papers

Speech-guided referring video object segmentation aims to recover the mask tracks of objects specified by a spoken motion description. Here, speech carries a linguistic instruction rather than acoustic evidence from a sounding object, so a solution must connect speech recognition, motion-centric temporal grounding, mask tracking, and explicit no-target handling. We introduce Speech2MaskTrack, our approach for the MeViS-Audio track of the 8th LSVOS Challenge. Speech2MaskTrack transcribes the spoken query and compiles it into structured constraints over category, count, direction, interaction role, and temporal phase. SAM3.1 enumerates multiple instance tracks, which TRACE ranks using complete-trajectory motion and relation evidence. A frozen lexical presence gate may suppress the ranked SAM3.1 base prediction. When the gate predicts that a target is present, an available full-expression-conditioned SaSaSa2VA track replaces the SAM3.1 mask. Only outputs that remain empty enter GPT-assisted recovery, which invokes SaSaSa2VA again under query- and mask-level verification. Speech2MaskTrack achieved second place in the official challenge ranking.

Multimodal Large Language Models (MLLMs) have achieved notable success in enhancing translation performance by integrating multimodal information. However, existing research primarily focuses on image-guided methods, whose applicability is constrained by the scarcity of multilingual image-text pairs. The speech modality overcomes this limitation due to its natural alignment with text and the abundance of existing speech datasets, which enable scalable language coverage. In this paper, we propose a Speech-guided Machine Translation (SMT) framework that integrates speech and text as fused inputs into an MLLM to improve translation quality. To mitigate reliance on low-resource data, we introduce a Self-Evolution Mechanism. The core components of this framework include a text-to-speech model, responsible for generating synthetic speech, and an MLLM capable of classifying synthetic speech samples and iteratively optimizing itself using positive samples. Experimental results demonstrate that our framework surpasses all existing methods on the Multi30K multimodal machine translation benchmark, achieving new state-of-the-art results. Furthermore, on general machine translation datasets, particularly the FLORES-200, it achieves average state-of-the-art performance in 108 translation directions. Ablation studies on CoVoST-2 confirms that differences between synthetic and authentic speech have negligible impact on translation quality. The code and models are released at https://github.com/yxduir/LLM-SRT.

The goal of this paper is to optimize the training process of diffusion-based text-to-speech models. While recent studies have achieved remarkable advancements, their training demands substantial time and computational costs, largely due to the implicit guidance of diffusion models in learning complex intermediate representations. To address this, we propose A-DMA, an effective strategy for Accelerating training with Dual Modality Alignment. Our method introduces a novel alignment pipeline leveraging both text and speech modalities: text-guided alignment, which incorporates contextual representations, and speech-guided alignment, which refines semantic representations. By aligning hidden states with discriminative features, our training scheme reduces the reliance on diffusion models for learning complex representations. Extensive experiments demonstrate that A-DMA doubles the convergence speed while achieving superior performance over baselines. Code and demo samples are available at: https://github.com/ZhikangNiu/A-DMA

Current dialogue generation approaches typically require the complete dialogue text before synthesis and produce a single, inseparable speech containing all voices, making them unsuitable for interactive chat; moreover, they suffer from unstable synthesis, inaccurate speaker transitions, and incoherent prosody. In this work, we present FireRedTTS-2, a long-form streaming TTS system for multi-speaker dialogue generation, delivering stable, natural speech with reliable speaker switching and context-aware prosody. A new 12.5Hz streaming speech tokenizer accelerates training and inference, extends maximum dialogue length, encodes richer semantics to stabilize text-to-token modeling and supports high-fidelity streaming generation for real-time applications. We adopt a text-speech interleaved format, concatenating speaker-labeled text with aligned speech tokens in chronological order, and model it with a dual-transformer: a large decoder-only transformer predicts tokens at the first layer, and a smaller one completes subsequent layers. Experimental results show that FireRedTTS-2 integrates seamlessly with chat frameworks and, with minimal fine-tuning, produces emotionally expressive speech guided by implicit contextual cues. In podcast generation, it surpasses existing systems including MoonCast, Zipvoice-Dialogue, and MOSS-TTSD in objective intelligibility, speaker-turn reliability, and perceived naturalness with context-consistent prosody. Our demos are available at https://fireredteam.github.io/demos/firered_tts_2.

We address the task of video chaptering, i.e., partitioning a long video timeline into semantic units and generating corresponding chapter titles. While relatively underexplored, automatic chaptering has the potential to enable efficient navigation and content retrieval in long-form videos. In this paper, we achieve strong chaptering performance on hour-long videos by efficiently addressing the problem in the text domain with our 'Chapter-Llama' framework. Specifically, we leverage a pretrained large language model (LLM) with large context window, and feed as input (i) speech transcripts and (ii) captions describing video frames, along with their respective timestamps. Given the inefficiency of exhaustively captioning all frames, we propose a lightweight speech-guided frame selection strategy based on speech transcript content, and experimentally demonstrate remarkable advantages. We train the LLM to output timestamps for the chapter boundaries, as well as free-form chapter titles. This simple yet powerful approach scales to processing one-hour long videos in a single forward pass. Our results demonstrate substantial improvements (e.g., 45.3 vs 26.7 F1 score) over the state of the art on the recent VidChapters-7M benchmark. To promote further research, we release our code and models at our project page.

Instruction-guided speech editing requires a model to modify specified speech attributes while preserving unrelated characteristics. Despite rapid progress in Speech Large Language Models (Speech LLMs), systematic evaluation of this capability remains challenging, as existing benchmarks are fragmented across isolated editing tasks. To bridge this gap, we introduce SpeechEditBench, a bilingual multi-attribute benchmark for instruction-guided speech editing. SpeechEditBench encompasses seven atomic editing tasks, as well as compositional editing tasks that integrate multiple operations within a single instruction. We propose an anchor-based evaluation protocol that separately assesses the edit success of target attributes and the preservation of untargeted attributes, leading to three metrics: target success, preservation success, and joint success. Using this benchmark, we evaluate mainstream Speech LLMs and specialized speech editing systems. The results reveal three key findings: (1) no single model performs well across all editing dimensions; (2) closed-source Speech LLMs generally outperform open-source models; (3) compositional editing remains highly challenging, with even the most advanced models struggling to achieve high joint success. SpeechEditBench provides a rigorous diagnostic framework to identify bottlenecks in Speech LLMs, thereby facilitating the development of next-generation Speech LLMs with more robust and precise instruction-guided editing capabilities. Data and code are avaialble at https://github.com/daxintan-cuhk/SpeechEditBench .

We propose Cotatron, a transcription-guided speech encoder for speaker-independent linguistic representation. Cotatron is based on the multispeaker TTS architecture and can be trained with conventional TTS datasets. We train a voice conversion system to reconstruct speech with Cotatron features, which is similar to the previous methods based on Phonetic Posteriorgram (PPG). By training and evaluating our system with 108 speakers from the VCTK dataset, we outperform the previous method in terms of both naturalness and speaker similarity. Our system can also convert speech from speakers that are unseen during training, and utilize ASR to automate the transcription with minimal reduction of the performance. Audio samples are available at https://mindslab-ai.github.io/cotatron, and the code with a pre-trained model will be made available soon.

The perceptual quality of neural text-to-speech (TTS) is highly dependent on the choice of the model during training. Selecting the model using a training-objective metric such as the least mean squared error does not always correlate with human perception. In this paper, we propose an objective metric based on the phone error rate (PER) to select the TTS model with the best speech intelligibility. The PER is computed between the input text to the TTS model, and the text decoded from the synthesized speech using an automatic speech recognition (ASR) model, which is trained on the same data as the TTS model. With the help of subjective studies, we show that the TTS model chosen with the least PER on validation split has significantly higher speech intelligibility compared to the model with the least training-objective metric loss. Finally, using the proposed PER and subjective evaluation, we show that the choice of best TTS model depends on the genre of the target domain text. All our experiments are conducted on a Hindi language dataset. However, the proposed model selection method is language independent.

Audio-visual alignment after dubbing is a challenging research problem. To this end, we propose a novel method, DubWise Multi-modal Large Language Model (LLM)-based Text-to-Speech (TTS), which can control the speech duration of synthesized speech in such a way that it aligns well with the speakers lip movements given in the reference video even when the spoken text is different or in a different language. To accomplish this, we propose to utilize cross-modal attention techniques in a pre-trained GPT-based TTS. We combine linguistic tokens from text, speaker identity tokens via a voice cloning network, and video tokens via a proposed duration controller network. We demonstrate the effectiveness of our system on the Lip2Wav-Chemistry and LRS2 datasets. Also, the proposed method achieves improved lip sync and naturalness compared to the SOTAs for the same language but different text (i.e., non-parallel) and the different language, different text (i.e., cross-lingual) scenarios.

Recent advancements in controllable expressive speech synthesis, especially in text-to-speech (TTS) models, have allowed for the generation of speech with specific styles guided by textual descriptions, known as style prompts. While this development enhances the flexibility and naturalness of synthesized speech, there remains a significant gap in understanding how these models handle vague or abstract style prompts. This study investigates the potential gender bias in how models interpret occupation-related prompts, specifically examining their responses to instructions like "Act like a nurse". We explore whether these models exhibit tendencies to amplify gender stereotypes when interpreting such prompts. Our experimental results reveal the model's tendency to exhibit gender bias for certain occupations. Moreover, models of different sizes show varying degrees of this bias across these occupations.

Overlapped speech is notoriously problematic for speaker diarization systems. Consequently, the use of speech separation has recently been proposed to improve their performance. Although promising, speech separation models struggle with realistic data because they are trained on simulated mixtures with a fixed number of speakers. In this work, we introduce a new speech separation-guided diarization scheme suitable for the online speaker diarization of long meeting recordings with a variable number of speakers, as present in the AMI corpus. We envisage ConvTasNet and DPRNN as alternatives for the separation networks, with two or three output sources. To obtain the speaker diarization result, voice activity detection is applied on each estimated source. The final model is fine-tuned end-to-end, after first adapting the separation to real data using AMI. The system operates on short segments, and inference is performed by stitching the local predictions using speaker embeddings and incremental clustering. The results show that our system improves the state-of-the-art on the AMI headset mix, using no oracle information and under full evaluation (no collar and including overlapped speech). Finally, we show the strength of our system particularly on overlapped speech sections.

Recent AIGC systems possess the capability to generate digital multimedia content based on human language instructions, such as text, image and video. However, when it comes to speech, existing methods related to human instruction-to-speech generation exhibit two limitations. Firstly, they require the division of inputs into content prompt (transcript) and description prompt (style and speaker), instead of directly supporting human instruction. This division is less natural in form and does not align with other AIGC models. Secondly, the practice of utilizing an independent description prompt to model speech style, without considering the transcript content, restricts the ability to control speech at a fine-grained level. To address these limitations, we propose VoxInstruct, a novel unified multilingual codec language modeling framework that extends traditional text-to-speech tasks into a general human instruction-to-speech task. Our approach enhances the expressiveness of human instruction-guided speech generation and aligns the speech generation paradigm with other modalities. To enable the model to automatically extract the content of synthesized speech from raw text instructions, we introduce speech semantic tokens as an intermediate representation for instruction-to-content guidance. We also incorporate multiple Classifier-Free Guidance (CFG) strategies into our codec language model, which strengthens the generated speech following human instructions. Furthermore, our model architecture and training strategies allow for the simultaneous support of combining speech prompt and descriptive human instruction for expressive speech synthesis, which is a first-of-its-kind attempt. Codes, models and demos are at: https://github.com/thuhcsi/VoxInstruct.

Existing speech models suffer from competing requirements on token representations by understanding and generation tasks. This discrepancy in representation prevents speech language models from performing instruction-based free-form editing. To solve this challenge, we introduce a novel framework that unifies speech understanding, generation, and editing. The core of our unified model is a unified continuous speech tokenizer MingTok-Audio, the first continuous tokenizer to effectively integrate semantic and acoustic features, which makes it suitable for both understanding and generation tasks. Based on this unified continuous audio tokenizer, we developed the speech language model Ming-UniAudio, which achieved a balance between generation and understanding capabilities. Ming-UniAudio sets new state-of-the-art (SOTA) records on 8 out of 12 metrics on the ContextASR benchmark. Notably, for Chinese voice cloning, it achieves a highly competitive Seed-TTS-WER of 0.95. Leveraging this foundational model, we further trained a dedicated speech editing model Ming-UniAudio-Edit, the first speech language model that enables universal, free-form speech editing guided solely by natural language instructions, handling both semantic and acoustic modifications without timestamp condition. To rigorously assess the editing capability and establish a foundation for future research, we introduce Ming-Freeform-Audio-Edit, the first comprehensive benchmark tailored for instruction-based free-form speech editing, featuring diverse scenarios and evaluation dimensions spanning semantic correctness, acoustic quality, and instruction alignment. We open-sourced the continuous audio tokenizer, the unified foundational model, and the free-form instruction-based editing model to facilitate the development of unified audio understanding, generation, and manipulation.

Controllable text-to-speech (TTS) has become a key research focus. However, methods based on either reference speech or text descriptions lack flexibility and precise control, and recent joint approaches remain loosely coupled, with speech modeling timbre and text controlling global style. We propose FineCombo-TTS, a unified framework for speech synthesis grounded in reference speech and guided by text descriptions, enabling flexible and precise control over acoustic attributes. Instead of explicit attribute disentanglement, we learn a unified acoustic representation and introduce a Conditional Flow Matching (CFM)-based Speech Variance Predictor to model fine-grained reference-to-target transformations guided by text descriptions. To support relative attribute control, we construct FineEdit, a structured paired dataset that explicitly encodes source-to-target attribute variations. Experiments demonstrate that our approach achieves flexible, precise, and expressive controllable TTS.

Classical TTS systems typically rely on rigid input formats and predefined metadata slots, limiting their ability to fulfill flexible user requirements. This paper introduces Bagpiper-TTS, a universal speech synthesis system that deals with diverse natural language user requests. Given a natural language prompt, Bagpiper-TTS first reasons over the users' intent to derive a rich caption, i.e., a comprehensive textual blueprint encompassing both transcription and nuanced metadata. Subsequently, this caption guides the synthesis of the target speech. Our model inherently supports a broad spectrum of tasks besides classical TTS applications, including multi-talker, intent-to-speech, role-play synthesis, singing voice synthesis, and more. Experimental results demonstrate that Bagpiper-TTS achieves an 1.7% Word Error Rate (WER) on the Seed-TTS-Eval benchmark and match the performance of dedicated models in both LLM-as-a-judge and human subjective evaluations across multiple applications.

Code-switch (CS) Automatic Speech Recognition (ASR) remains challenging due to limited availability of high quality CS text-speech pairs for training. Although synthetic data augmentation via Text-to-speech (TTS) has been explored, existing CS TTS approaches primarily optimise reconstruction fidelity and do not explicitly enforce language-boundary consistency, thereby limiting their effectiveness for CS ASR augmentation. This paper proposes a code-mixing guided preference-learning framework that steers synthetic speech generation toward improved code-switching fidelity using the Code Mixing Index (CMI). Experiments on the SEAME Mandarin-English conversational corpus demonstrate that the proposed method enhances the utility of synthetic data for ASR fine-tuning. Specifically, when fine-tuning Whisper Large, the proposed approach reduces Mixed Error Rate (MER) from 12.1%/17.8% to 8.9%/14.2% on the DevMAN and DevSGE sets, respectively.

We introduce Baichuan-Audio, an end-to-end audio large language model that seamlessly integrates audio understanding and generation. It features a text-guided aligned speech generation mechanism, enabling real-time speech interaction with both comprehension and generation capabilities. Baichuan-Audio leverages a pre-trained ASR model, followed by multi-codebook discretization of speech at a frame rate of 12.5 Hz. This multi-codebook setup ensures that speech tokens retain both semantic and acoustic information. To further enhance modeling, an independent audio head is employed to process audio tokens, effectively capturing their unique characteristics. To mitigate the loss of intelligence during pre-training and preserve the original capabilities of the LLM, we propose a two-stage pre-training strategy that maintains language understanding while enhancing audio modeling. Following alignment, the model excels in real-time speech-based conversation and exhibits outstanding question-answering capabilities, demonstrating its versatility and efficiency. The proposed model demonstrates superior performance in real-time spoken dialogue and exhibits strong question-answering abilities. Our code, model and training data are available at https://github.com/baichuan-inc/Baichuan-Audio

Recent advancements in speech-language models have yielded significant improvements in speech tokenization and synthesis. However, effectively mapping the complex, multidimensional attributes of speech into discrete tokens remains challenging. This process demands acoustic, semantic, and contextual information for precise speech representations. Existing speech representations generally fall into two categories: acoustic tokens from audio codecs and semantic tokens from speech self-supervised learning models. Although recent efforts have unified acoustic and semantic tokens for improved performance, they overlook the crucial role of contextual representation in comprehensive speech modeling. Our empirical investigations reveal that the absence of contextual representations results in elevated Word Error Rate (WER) and Word Information Lost (WIL) scores in speech transcriptions. To address these limitations, we propose two novel distillation approaches: (1) a language model (LM)-guided distillation method that incorporates contextual information, and (2) a combined LM and self-supervised speech model (SM)-guided distillation technique that effectively distills multimodal representations (acoustic, semantic, and contextual) into a comprehensive speech tokenizer, termed DM-Codec. The DM-Codec architecture adopts a streamlined encoder-decoder framework with a Residual Vector Quantizer (RVQ) and incorporates the LM and SM during the training process. Experiments show DM-Codec significantly outperforms state-of-the-art speech tokenization models, reducing WER by up to 13.46%, WIL by 9.82%, and improving speech quality by 5.84% and intelligibility by 1.85% on the LibriSpeech benchmark dataset. The code, samples, and model checkpoints are available at https://github.com/mubtasimahasan/DM-Codec.

Existing Existing automatic audio generation methods struggle to generate podcast-like audio programs effectively. The key challenges lie in in-depth content generation, appropriate and expressive voice production. This paper proposed PodAgent, a comprehensive framework for creating audio programs. PodAgent 1) generates informative topic-discussion content by designing a Host-Guest-Writer multi-agent collaboration system, 2) builds a voice pool for suitable voice-role matching and 3) utilizes LLM-enhanced speech synthesis method to generate expressive conversational speech. Given the absence of standardized evaluation criteria for podcast-like audio generation, we developed comprehensive assessment guidelines to effectively evaluate the model's performance. Experimental results demonstrate PodAgent's effectiveness, significantly surpassing direct GPT-4 generation in topic-discussion dialogue content, achieving an 87.4% voice-matching accuracy, and producing more expressive speech through LLM-guided synthesis. Demo page: https://podcast-agent.github.io/demo/. Source code: https://github.com/yujxx/PodAgent.

Masked speech modeling (MSM) methods such as wav2vec2 or w2v-BERT learn representations over speech frames which are randomly masked within an utterance. While these methods improve performance of Automatic Speech Recognition (ASR) systems, they have one major limitation. They treat all unsupervised speech samples with equal weight, which hinders learning as not all samples have relevant information to learn meaningful representations. In this work, we address this limitation. We propose ask2mask (ATM), a novel approach to focus on specific samples during MSM pre-training. ATM employs an external ASR model or scorer to weight unsupervised input samples in two different ways: 1) A fine-grained data selection is performed by masking over the highly confident input frames as chosen by the scorer. This allows the model to learn meaningful representations. 2) ATM is further extended to focus at utterance-level by weighting the final MSM loss with the utterance-level confidence score. We conduct fine-tuning experiments on two well-benchmarked corpora: LibriSpeech (matching the pre-training data) and Commonvoice, TED-LIUM, AMI and CHiME-6 (not matching the pre-training data). The results substantiate the efficacy of ATM on significantly improving the recognition performance under mismatched conditions (up to 11.6\% relative over published results and upto 4.46\% relative over our internal baseline) while still yielding modest improvements under matched conditions.

Speech-driven 3D motion synthesis seeks to create lifelike animations based on human speech, with potential uses in virtual reality, gaming, and the film production. Existing approaches reply solely on speech audio for motion generation, leading to inaccurate and inflexible synthesis results. To mitigate this problem, we introduce a novel text-guided 3D human motion synthesis method, termed T3M. Unlike traditional approaches, T3M allows precise control over motion synthesis via textual input, enhancing the degree of diversity and user customization. The experiment results demonstrate that T3M can greatly outperform the state-of-the-art methods in both quantitative metrics and qualitative evaluations. We have publicly released our code at https://github.com/Gloria2tt/T3M.git{https://github.com/Gloria2tt/T3M.git}

Social media platforms, despite their value in promoting open discourse, are often exploited to spread harmful content. Current deep learning and natural language processing models used for detecting this harmful content overly rely on domain-specific terms affecting their capabilities to adapt to generalizable hate speech detection. This is because they tend to focus too narrowly on particular linguistic signals or the use of certain categories of words. Another significant challenge arises when platforms lack high-quality annotated data for training, leading to a need for cross-platform models that can adapt to different distribution shifts. Our research introduces a cross-platform hate speech detection model capable of being trained on one platform's data and generalizing to multiple unseen platforms. To achieve good generalizability across platforms, one way is to disentangle the input representations into invariant and platform-dependent features. We also argue that learning causal relationships, which remain constant across diverse environments, can significantly aid in understanding invariant representations in hate speech. By disentangling input into platform-dependent features (useful for predicting hate targets) and platform-independent features (used to predict the presence of hate), we learn invariant representations resistant to distribution shifts. These features are then used to predict hate speech across unseen platforms. Our extensive experiments across four platforms highlight our model's enhanced efficacy compared to existing state-of-the-art methods in detecting generalized hate speech.

Instruction-guided text-to-speech (ITTS) enables users to control speech generation through natural language prompts, offering a more intuitive interface than traditional TTS. However, the alignment between user style instructions and listener perception remains largely unexplored. This work first presents a perceptual analysis of ITTS controllability across two expressive dimensions (adverbs of degree and graded emotion intensity) and collects human ratings on speaker age and word-level emphasis attributes. To comprehensively reveal the instruction-perception gap, we provide a data collection with large-scale human evaluations, named Expressive VOice Control (E-VOC) corpus. Furthermore, we reveal that (1) gpt-4o-mini-tts is the most reliable ITTS model with great alignment between instruction and generated utterances across acoustic dimensions. (2) The 5 analyzed ITTS systems tend to generate Adult voices even when the instructions ask to use child or Elderly voices. (3) Fine-grained control remains a major challenge, indicating that most ITTS systems have substantial room for improvement in interpreting slightly different attribute instructions.

The proliferation of online hate speech poses a significant threat to the harmony of the web. While explicit hate is easily recognized through overt slurs, implicit hate speech is often conveyed through sarcasm, irony, stereotypes, or coded language -- making it harder to detect. Existing hate speech detection models, which predominantly rely on surface-level linguistic cues, fail to generalize effectively across diverse stylistic variations. Moreover, hate speech spread on different platforms often targets distinct groups and adopts unique styles, potentially inducing spurious correlations between them and labels, further challenging current detection approaches. Motivated by these observations, we hypothesize that the generation of hate speech can be modeled as a causal graph involving key factors: contextual environment, creator motivation, target, and style. Guided by this graph, we propose CADET, a causal representation learning framework that disentangles hate speech into interpretable latent factors and then controls confounders, thereby isolating genuine hate intent from superficial linguistic cues. Furthermore, CADET allows counterfactual reasoning by intervening on style within the latent space, naturally guiding the model to robustly identify hate speech in varying forms. CADET demonstrates superior performance in comprehensive experiments, highlighting the potential of causal priors in advancing generalizable hate speech detection.

Self supervised representation learning has recently attracted a lot of research interest for both the audio and visual modalities. However, most works typically focus on a particular modality or feature alone and there has been very limited work that studies the interaction between the two modalities for learning self supervised representations. We propose a framework for learning audio representations guided by the visual modality in the context of audiovisual speech. We employ a generative audio-to-video training scheme in which we animate a still image corresponding to a given audio clip and optimize the generated video to be as close as possible to the real video of the speech segment. Through this process, the audio encoder network learns useful speech representations that we evaluate on emotion recognition and speech recognition. We achieve state of the art results for emotion recognition and competitive results for speech recognition. This demonstrates the potential of visual supervision for learning audio representations as a novel way for self-supervised learning which has not been explored in the past. The proposed unsupervised audio features can leverage a virtually unlimited amount of training data of unlabelled audiovisual speech and have a large number of potentially promising applications.

Large-scale generative models such as GPT and DALL-E have revolutionized the research community. These models not only generate high fidelity outputs, but are also generalists which can solve tasks not explicitly taught. In contrast, speech generative models are still primitive in terms of scale and task generalization. In this paper, we present Voicebox, the most versatile text-guided generative model for speech at scale. Voicebox is a non-autoregressive flow-matching model trained to infill speech, given audio context and text, trained on over 50K hours of speech that are not filtered or enhanced. Similar to GPT, Voicebox can perform many different tasks through in-context learning, but is more flexible as it can also condition on future context. Voicebox can be used for mono or cross-lingual zero-shot text-to-speech synthesis, noise removal, content editing, style conversion, and diverse sample generation. In particular, Voicebox outperforms the state-of-the-art zero-shot TTS model VALL-E on both intelligibility (5.9% vs 1.9% word error rates) and audio similarity (0.580 vs 0.681) while being up to 20 times faster. Audio samples can be found in https://voicebox.metademolab.com.

Multi-speaker text-to-speech (TTS) using a few adaption data is a challenge in practical applications. To address that, we propose a zero-shot multi-speaker TTS, named nnSpeech, that could synthesis a new speaker voice without fine-tuning and using only one adaption utterance. Compared with using a speaker representation module to extract the characteristics of new speakers, our method bases on a speaker-guided conditional variational autoencoder and can generate a variable Z, which contains both speaker characteristics and content information. The latent variable Z distribution is approximated by another variable conditioned on reference mel-spectrogram and phoneme. Experiments on the English corpus, Mandarin corpus, and cross-dataset proves that our model could generate natural and similar speech with only one adaption speech.

Hate speech detection refers to the task of detecting hateful content that aims at denigrating an individual or a group based on their religion, gender, sexual orientation, or other characteristics. Due to the different policies of the platforms, different groups of people express hate in different ways. Furthermore, due to the lack of labeled data in some platforms it becomes challenging to build hate speech detection models. To this end, we revisit if we can learn a generalizable hate speech detection model for the cross platform setting, where we train the model on the data from one (source) platform and generalize the model across multiple (target) platforms. Existing generalization models rely on linguistic cues or auxiliary information, making them biased towards certain tags or certain kinds of words (e.g., abusive words) on the source platform and thus not applicable to the target platforms. Inspired by social and psychological theories, we endeavor to explore if there exist inherent causal cues that can be leveraged to learn generalizable representations for detecting hate speech across these distribution shifts. To this end, we propose a causality-guided framework, PEACE, that identifies and leverages two intrinsic causal cues omnipresent in hateful content: the overall sentiment and the aggression in the text. We conduct extensive experiments across multiple platforms (representing the distribution shift) showing if causal cues can help cross-platform generalization.

We propose Guided-TTS, a high-quality text-to-speech (TTS) model that does not require any transcript of target speaker using classifier guidance. Guided-TTS combines an unconditional diffusion probabilistic model with a separately trained phoneme classifier for classifier guidance. Our unconditional diffusion model learns to generate speech without any context from untranscribed speech data. For TTS synthesis, we guide the generative process of the diffusion model with a phoneme classifier trained on a large-scale speech recognition dataset. We present a norm-based scaling method that reduces the pronunciation errors of classifier guidance in Guided-TTS. We show that Guided-TTS achieves a performance comparable to that of the state-of-the-art TTS model, Grad-TTS, without any transcript for LJSpeech. We further demonstrate that Guided-TTS performs well on diverse datasets including a long-form untranscribed dataset.

This paper describes our self-designed system for Task 1 of the MLC-SLM 2026 Challenge for multilingual two-speaker conversational speech. The system combines a modular speaker diarization front end with a challenge-adapted Qwen3-ASR-1.7B recognizer. The diarization front end performs voice activity detection, subsegment generation, CAMPPlus speaker embedding extraction, two-speaker spectral clustering, and RTTM-based audio segmentation. The resulting speaker-attributed segments are grouped by language or region and decoded by the adapted ASR model. For ASR adaptation, we first perform supervised full fine-tuning on the official training data, then apply LoRA fine-tuning with synthetic speech generated by a three-pipeline TTS-based synthetic speech augmentation framework, and finally refine the model using GRPO reinforcement learning with rewards based on WER/CER and penalties for hallucination, repetition, and length deviation. On the official development set, the full system achieves an average tcpMER of 23.70, reducing the error rate by 6.83 absolute points relative to the released Qwen-ASR-1.7B performance. On the final evaluation set, the system achieves an average tcpMER of 17.97. Ablation results show that supervised fine-tuning provides the largest gain, while synthetic-speech LoRA adaptation and reinforcement learning further improve robustness.

Recent research has demonstrated impressive results in video-to-speech synthesis which involves reconstructing speech solely from visual input. However, previous works have struggled to accurately synthesize speech due to a lack of sufficient guidance for the model to infer the correct content with the appropriate sound. To resolve the issue, they have adopted an extra speaker embedding as a speaking style guidance from a reference auditory information. Nevertheless, it is not always possible to obtain the audio information from the corresponding video input, especially during the inference time. In this paper, we present a novel vision-guided speaker embedding extractor using a self-supervised pre-trained model and prompt tuning technique. In doing so, the rich speaker embedding information can be produced solely from input visual information, and the extra audio information is not necessary during the inference time. Using the extracted vision-guided speaker embedding representations, we further develop a diffusion-based video-to-speech synthesis model, so called DiffV2S, conditioned on those speaker embeddings and the visual representation extracted from the input video. The proposed DiffV2S not only maintains phoneme details contained in the input video frames, but also creates a highly intelligible mel-spectrogram in which the speaker identities of the multiple speakers are all preserved. Our experimental results show that DiffV2S achieves the state-of-the-art performance compared to the previous video-to-speech synthesis technique.

Lip-to-speech involves generating a natural-sounding speech synchronized with a soundless video of a person talking. Despite recent advances, current methods still cannot produce high-quality speech with high levels of intelligibility for challenging and realistic datasets such as LRS3. In this work, we present LipVoicer, a novel method that generates high-quality speech, even for in-the-wild and rich datasets, by incorporating the text modality. Given a silent video, we first predict the spoken text using a pre-trained lip-reading network. We then condition a diffusion model on the video and use the extracted text through a classifier-guidance mechanism where a pre-trained ASR serves as the classifier. LipVoicer outperforms multiple lip-to-speech baselines on LRS2 and LRS3, which are in-the-wild datasets with hundreds of unique speakers in their test set and an unrestricted vocabulary. Moreover, our experiments show that the inclusion of the text modality plays a major role in the intelligibility of the produced speech, readily perceptible while listening, and is empirically reflected in the substantial reduction of the WER metric. We demonstrate the effectiveness of LipVoicer through human evaluation, which shows that it produces more natural and synchronized speech signals compared to competing methods. Finally, we created a demo showcasing LipVoicer's superiority in producing natural, synchronized, and intelligible speech, providing additional evidence of its effectiveness. Project page and code: https://github.com/yochaiye/LipVoicer

This paper describes a novel text-to-speech (TTS) technique based on deep convolutional neural networks (CNN), without use of any recurrent units. Recurrent neural networks (RNN) have become a standard technique to model sequential data recently, and this technique has been used in some cutting-edge neural TTS techniques. However, training RNN components often requires a very powerful computer, or a very long time, typically several days or weeks. Recent other studies, on the other hand, have shown that CNN-based sequence synthesis can be much faster than RNN-based techniques, because of high parallelizability. The objective of this paper is to show that an alternative neural TTS based only on CNN alleviate these economic costs of training. In our experiment, the proposed Deep Convolutional TTS was sufficiently trained overnight (15 hours), using an ordinary gaming PC equipped with two GPUs, while the quality of the synthesized speech was almost acceptable.

Invasive brain-computer interfaces have garnered significant attention due to their high performance. The current intracranial stereoElectroEncephaloGraphy (sEEG) foundation models typically build univariate representations based on a single channel. Some of them further use Transformer to model the relationship among channels. However, due to the locality and specificity of brain computation, their performance on more difficult tasks, e.g., speech decoding, which demands intricate processing in specific brain regions, is yet to be fully investigated. We hypothesize that building multi-variate representations within certain brain regions can better capture the specific neural processing. To explore this hypothesis, we collect a well-annotated Chinese word-reading sEEG dataset, targeting language-related brain networks, over 12 subjects. Leveraging this benchmark dataset, we developed the Du-IN model that can extract contextual embeddings from specific brain regions through discrete codebook-guided mask modeling. Our model achieves SOTA performance on the downstream 61-word classification task, surpassing all baseline models. Model comparison and ablation analysis reveal that our design choices, including (i) multi-variate representation by fusing channels in vSMC and STG regions and (ii) self-supervision by discrete codebook-guided mask modeling, significantly contribute to these performances. Collectively, our approach, inspired by neuroscience findings, capitalizing on multi-variate neural representation from specific brain regions, is suitable for invasive brain modeling. It marks a promising neuro-inspired AI approach in BCI.

We propose Guided-TTS 2, a diffusion-based generative model for high-quality adaptive TTS using untranscribed data. Guided-TTS 2 combines a speaker-conditional diffusion model with a speaker-dependent phoneme classifier for adaptive text-to-speech. We train the speaker-conditional diffusion model on large-scale untranscribed datasets for a classifier-free guidance method and further fine-tune the diffusion model on the reference speech of the target speaker for adaptation, which only takes 40 seconds. We demonstrate that Guided-TTS 2 shows comparable performance to high-quality single-speaker TTS baselines in terms of speech quality and speaker similarity with only a ten-second untranscribed data. We further show that Guided-TTS 2 outperforms adaptive TTS baselines on multi-speaker datasets even with a zero-shot adaptation setting. Guided-TTS 2 can adapt to a wide range of voices only using untranscribed speech, which enables adaptive TTS with the voice of non-human characters such as Gollum in "The Lord of the Rings".

Using unsupervised learning to disentangle speech into content, rhythm, pitch, and timbre for voice conversion has become a hot research topic. Existing works generally take into account disentangling speech components through human-crafted bottleneck features which can not achieve sufficient information disentangling, while pitch and rhythm may still be mixed together. There is a risk of information overlap in the disentangling process which results in less speech naturalness. To overcome such limits, we propose a two-stage model to disentangle speech representations in a self-supervised manner without a human-crafted bottleneck design, which uses the Mutual Information (MI) with the designed upper bound estimator (IFUB) to separate overlapping information between speech components. Moreover, we design a Joint Text-Guided Consistent (TGC) module to guide the extraction of speech content and eliminate timbre leakage issues. Experiments show that our model can achieve a better performance than the baseline, regarding disentanglement effectiveness, speech naturalness, and similarity. Audio samples can be found at https://largeaudiomodel.com/eadvc.

Recent alignment-free non-autoregressive (NAR) text-to-speech (TTS) models formulate synthesis as a conditional infilling task, bypassing explicit duration predictors and external aligners. When speech is represented with neural codec tokens, the infilling problem becomes discrete, making Discrete Flow Matching (DFM), a Continuous-Time Markov Chain (CTMC) framework for discrete generation, a natural fit. However, inference-time control for stable low-step conditional infilling remains underexplored. We propose Mask, Sample, Revise, an inference-time CTMC stack for alignment-free DFM-TTS. The stack combines predictor-free guidance to strengthen text conditioning, prompt-matched conditional coupling to align the probability path with the acoustic prompt, and SC-ReMask, a schedule-constrained remasking mechanism that introduces token-to-mask transitions so early de-masking decisions can be revised. These components require no post-hoc fine-tuning and operate in a single tau-leaping sampler. Controlled ablations show that this stack improves intelligibility and robustness in the low-NFE prompted setting, outperforming unguided and guidance-only samplers with substantially more steps.

Speech quality assessment (SQA) aims to predict the perceived quality of speech signals under a wide range of distortions. It is inherently connected to speech enhancement (SE), which seeks to improve speech quality by removing unwanted signal components. While SQA models are widely used to evaluate SE performance, their potential to guide SE training remains underexplored. In this work, we investigate a training framework that leverages a SQA model, trained to predict multiple evaluation metrics from a public SE leaderboard, as a supervisory signal for SE. This approach addresses a key limitation of conventional SE objectives, such as SI-SNR, which often fail to align with perceptual quality and generalize poorly across evaluation metrics. Moreover, it enables training on real-world data where clean references are unavailable. Experiments on both simulated and real-world test sets show that SQA-guided training consistently improves performance across a range of quality metrics. Code and checkpoints are available at https://github.com/urgent-challenge/urgent2026_challenge_track2

The pre-trained speech encoder wav2vec 2.0 performs very well on various spoken language understanding (SLU) tasks. However, on many tasks, it trails behind text encoders with textual input. To improve the understanding capability of SLU encoders, various studies have used knowledge distillation to transfer knowledge from natural language understanding (NLU) encoders. We use a very simple method of distilling from a textual sentence embedder directly into wav2vec 2.0 as pre-training, utilizing paired audio-text datasets. We observed that this method is indeed capable of improving SLU task performance in fine-tuned settings, as well as full-data and few-shot transfer on a frozen encoder. However, the model performs worse on certain tasks highlighting the strengths and weaknesses of our approach.

Speech-driven 3D face animation aims to generate realistic facial expressions that match the speech content and emotion. However, existing methods often neglect emotional facial expressions or fail to disentangle them from speech content. To address this issue, this paper proposes an end-to-end neural network to disentangle different emotions in speech so as to generate rich 3D facial expressions. Specifically, we introduce the emotion disentangling encoder (EDE) to disentangle the emotion and content in the speech by cross-reconstructed speech signals with different emotion labels. Then an emotion-guided feature fusion decoder is employed to generate a 3D talking face with enhanced emotion. The decoder is driven by the disentangled identity, emotional, and content embeddings so as to generate controllable personal and emotional styles. Finally, considering the scarcity of the 3D emotional talking face data, we resort to the supervision of facial blendshapes, which enables the reconstruction of plausible 3D faces from 2D emotional data, and contribute a large-scale 3D emotional talking face dataset (3D-ETF) to train the network. Our experiments and user studies demonstrate that our approach outperforms state-of-the-art methods and exhibits more diverse facial movements. We recommend watching the supplementary video: https://ziqiaopeng.github.io/emotalk

Existing text scoring methods require a large corpus, struggle with short texts, or require hand-labeled data. We develop a text scoring framework that leverages generative large language models (LLMs) to (1) set texts against the backdrop of information from the near-totality of the web and digitized media, and (2) effectively transform pairwise text comparisons from a reasoning problem to a pattern recognition task. Our approach, concept-guided chain-of-thought (CGCoT), utilizes a chain of researcher-designed prompts with an LLM to generate a concept-specific breakdown for each text, akin to guidance provided to human coders. We then pairwise compare breakdowns using an LLM and aggregate answers into a score using a probability model. We apply this approach to better understand speech reflecting aversion to specific political parties on Twitter, a topic that has commanded increasing interest because of its potential contributions to democratic backsliding. We achieve stronger correlations with human judgments than widely used unsupervised text scoring methods like Wordfish. In a supervised setting, besides a small pilot dataset to develop CGCoT prompts, our measures require no additional hand-labeled data and produce predictions on par with RoBERTa-Large fine-tuned on thousands of hand-labeled tweets. This project showcases the potential of combining human expertise and LLMs for scoring tasks.

Large Language Model (LLM) based text-to-speech (TTS) systems have demonstrated remarkable capabilities in handling large speech datasets and generating natural speech for new speakers. However, LLM-based TTS models are not robust as the generated output can contain repeating words, missing words and mis-aligned speech (referred to as hallucinations or attention errors), especially when the text contains multiple occurrences of the same token. We examine these challenges in an encoder-decoder transformer model and find that certain cross-attention heads in such models implicitly learn the text and speech alignment when trained for predicting speech tokens for a given text. To make the alignment more robust, we propose techniques utilizing CTC loss and attention priors that encourage monotonic cross-attention over the text tokens. Our guided attention training technique does not introduce any new learnable parameters and significantly improves robustness of LLM-based TTS models.

Neural codec language models enable high-quality discrete speech synthesis, yet their inference remains vulnerable to token-level artifacts and distributional drift that degrade perceptual realism. Rather than relying on preference optimization or retraining, we propose MSpoof-TTS, a training-free inference framework that improves zero-shot synthesis through multi-resolution spoof guidance. We introduce a Multi-Resolution Token-based Spoof Detection framework that evaluates codec sequences at different temporal granularities to detect locally inconsistent or unnatural patterns. We then integrate the spoof detectors into a hierarchical decoding strategy, progressively pruning low-quality candidates and re-ranking hypotheses. This discriminator-guided generation enhances robustness without modifying model parameters. Experiments validate the effectiveness of our framework for robust and high-quality codec-based speech generation. Audio samples are available at https://danny-nus.github.io/MSpoofTTS.github.io/.

With the rapid advancement of speech generation technologies, the threat posed by speech deepfakes in real-time communication (RTC) scenarios has intensified. However, existing detection studies mainly focus on offline simulations and struggle to cope with the complex distortions introduced during RTC transmission, including unknown speech enhancement processes (e.g., noise suppression) and codec compression. To address this challenge, we present the first large-scale speech deepfake dataset tailored for RTC scenarios, termed RTCFake, totaling approximately 600 hours. The dataset is constructed by transmitting speech through multiple mainstream social media and conferencing platforms (e.g., Zoom), enabling precise pairing between offline and online speech. In addition, we propose a phoneme-guided consistency learning (PCL) strategy that enforces models to learn platform-invariant semantic structural representations. In this paper, the RTCFake dataset is divided into training, development, and evaluation sets. The evaluation set further includes both unseen RTC platforms and unseen complex noise conditions, thereby providing a more realistic and challenging evaluation benchmark for speech deepfake detection. Furthermore, the proposed PCL strategy achieves significant improvements in both cross-platform generalization and noise robustness, offering an effective and generalizable modeling paradigm. The RTCFake dataset is provided in the {https://huggingface.co/datasets/JunXueTech/RTCFake}.

Real-time co-speech gesture generation must produce 3D motion clip by clip as speech arrives. Existing streaming methods are open-loop: each clip depends on past context, but the model cannot check or correct its trajectory. Small errors therefore accumulate and cause drift over long sequences. We observe that this failure is mainly caused by the lack of a forward constraint rather than poor short-clip quality. A plausible key pose at the end of each clip provides a destination anchor that limits drift. Based on this observation, we propose StreamTalk, a closed-loop framework with a periodic generate-retrieve-refine cycle. Streaming Pose-Guided Generation first predicts a coarse clip, retrieves a plausible tail pose from a speaker-specific motion database, and refines the clip using this pose before continuing to the next window. During training, Stochastic Anchor Masking randomly masks pose and translation frames, teaching the model to recover complete motion from sparse boundary conditions. A part-aware DiT separates hand, body, and translation streams to reduce interference between global displacement and local articulation. On BEAT2, StreamTalk achieves state-of-the-art FGD, reduces long-horizon drift relative to open-loop baselines, and runs in real time at 76 FPS. Project page: https://xiangyue-zhang.github.io/StreamTalk/.

Neural speech codecs face a fundamental tension in the language-model era: tokens that support high-fidelity reconstruction are not necessarily easy for autoregressive models to predict. Our controlled analysis of diverse codec and self-supervised speech representations shows that clearer phoneme structure before discrete code assignment is consistently associated with easier autoregressive token prediction. Yet phoneme structure alone is insufficient for high-fidelity reconstruction, which also requires reconstruction-relevant acoustic detail. Guided by this observation, we introduce ReLMCodec, a low-bitrate single-codebook speech codec built upon a preserve--control--refine principle: it preserves the linguistic organization of frozen self-supervised learning (SSL) features at the quantizer input, controls reconstruction-driven drift through Pre-quantization Anchor-Preserving Adaptation (PAPA), and refines the quantized latent space with a training-only WavLM-Large L24 teacher to reduce phoneme-level token fragmentation. Together, these components allow acoustic detail to support waveform reconstruction while keeping the resulting token sequence predictable for autoregressive models. At 650 and 800 bps, ReLMCodec moves the empirical single-stream predictability--reconstruction frontier in our evaluations, with gains that carry over to downstream text-to-speech (TTS) synthesis in both intelligibility and speaker similarity.

Zero-shot text-to-speech (TTS) can reproduce an unseen speaker from a short reference recording, but typically entangles speaker identity and accent within the same reference. We introduce DEFINE, an end-to-end framework that decouples these factors by conditioning speaker identity and target accent on separate audio exemplars. A single inference-time guidance weight continuously controls accent strength without retraining. Built on F5-TTS with parameter-efficient LoRA adaptation, DEFINE maps short accent exemplars into a conditioning space using an exemplar encoder supervised through learned accent prototypes, requiring neither accent labels at inference time nor post-synthesis waveform conversion. On seen accents, increasing accent guidance improves accent-probe accuracy from 6.5% to 19.6%. More importantly, a single DEFINE model generalizes accent control beyond its training accent set: on seen and out-of-domain accents, though not on held-out accents, it matches the accent transfer performance of a two-model TTS-voice-conversion cascade while achieving higher speaker similarity and comparable predicted speech quality. These results demonstrate that speaker identity and accent can be independently controlled from audio exemplars within a single zero-shot TTS model, including for accents unseen during training.

Multimodal question answering (QA) often requires identifying which video, audio, or sensor tokens are relevant to the question. Yet modality disagreements are common: off-camera speech, background noise, or motion outside the field of view often mislead fusion models that weight all streams equally. We present RAVEN, a unified QA architecture whose core is QuART, a query-conditioned cross-modal gating module that assigns scalar relevance scores to each token across modalities, enabling the model to amplify informative signals and suppress distractors before fusion. RAVEN is trained through a three-stage pipeline comprising unimodal pretraining, query-aligned fusion, and disagreement-oriented fine-tuning -- each stage targeting a distinct challenge in multi-modal reasoning: representation quality, cross-modal relevance, and robustness to modality mismatch. To support training and evaluation, we release AVS-QA, a dataset of 300K synchronized Audio--Video-Sensor streams paired with automatically generated question-answer pairs. Experimental results on seven multi-modal QA benchmarks -- including egocentric and exocentric tasks -- show that RAVEN achieves up to 14.5\% and 8.0\% gains in accuracy compared to state-of-the-art multi-modal large language models, respectively. Incorporating sensor data provides an additional 16.4\% boost, and the model remains robust under modality corruption, outperforming SOTA baselines by 50.23\%. Our code and dataset are available at https://github.com/BASHLab/RAVEN.

Spoken dialogue systems often rely on cascaded pipelines that transcribe, process, and resynthesize speech. While effective, this design discards paralinguistic cues and limits expressivity. Recent end-to-end methods reduce latency and better preserve these cues, yet still rely on text intermediates, creating a fundamental bottleneck. We present MOSS-Speech, a true speech-to-speech large language model that directly understands and generates speech without relying on text guidance. Our approach combines a modality-based layer-splitting architecture with a frozen pre-training strategy, preserving the reasoning and knowledge of pretrained text LLMs while adding native speech capabilities. Experiments show that our model achieves state-of-the-art results in spoken question answering and delivers comparable speech-to-speech performance relative to existing text-guided systems, while still maintaining competitive text performance. By narrowing the gap between text-guided and direct speech generation, our work establishes a new paradigm for expressive and efficient end-to-end speech interaction.

This study presents a novel zero-shot user-defined keyword spotting model that utilizes the audio-phoneme relationship of the keyword to improve performance. Unlike the previous approach that estimates at utterance level, we use both utterance and phoneme level information. Our proposed method comprises a two-stream speech encoder architecture, self-attention-based pattern extractor, and phoneme-level detection loss for high performance in various pronunciation environments. Based on experimental results, our proposed model outperforms the baseline model and achieves competitive performance compared with full-shot keyword spotting models. Our proposed model significantly improves the EER and AUC across all datasets, including familiar words, proper nouns, and indistinguishable pronunciations, with an average relative improvement of 67% and 80%, respectively. The implementation code of our proposed model is available at https://github.com/ncsoft/PhonMatchNet.

As a robust and large-scale multilingual speech recognition model, Whisper has demonstrated impressive results in many low-resource and out-of-distribution scenarios. However, its encoder-decoder structure hinders its application to streaming speech recognition. In this paper, we introduce Simul-Whisper, which uses the time alignment embedded in Whisper's cross-attention to guide auto-regressive decoding and achieve chunk-based streaming ASR without any fine-tuning of the pre-trained model. Furthermore, we observe the negative effect of the truncated words at the chunk boundaries on the decoding results and propose an integrate-and-fire-based truncation detection model to address this issue. Experiments on multiple languages and Whisper architectures show that Simul-Whisper achieves an average absolute word error rate degradation of only 1.46% at a chunk size of 1 second, which significantly outperforms the current state-of-the-art baseline.

Generative models have shown remarkable performance in speech enhancement (SE), achieving superior perceptual quality over traditional discriminative approaches. However, existing generative SE approaches often overlook the risk of hallucination under severe noise, leading to incorrect spoken content or inconsistent speaker characteristics, which we term linguistic and acoustic hallucinations, respectively. We argue that linguistic hallucination stems from models' failure to constrain valid phonological structures and it is a more fundamental challenge. While language models (LMs) are well-suited for capturing the underlying speech structure through modeling the distribution of discrete tokens, existing approaches are limited in learning from noise-corrupted representations, which can lead to contaminated priors and hallucinations. To overcome these limitations, we propose the Phonologically Anchored Speech Enhancer (PASE), a generative SE framework that leverages the robust phonological prior embedded in the pre-trained WavLM model to mitigate hallucinations. First, we adapt WavLM into a denoising expert via representation distillation to clean its final-layer features. Guided by the model's intrinsic phonological prior, this process enables robust denoising while minimizing linguistic hallucinations. To further reduce acoustic hallucinations, we train the vocoder with a dual-stream representation: the high-level phonetic representation provides clean linguistic content, while a low-level acoustic representation retains speaker identity and prosody. Experimental results demonstrate that PASE not only surpasses state-of-the-art discriminative models in perceptual quality, but also significantly outperforms prior generative models with substantially lower linguistic and acoustic hallucinations.

Recent advancements in text-guided audio generation have yielded promising results in diverse domains, including sound effects, speech, and music. However, jointly generating speech with environmental audio remains challenging due to the inherent disparities in their acoustic patterns and temporal dynamics. We propose ImmersiveTTS, an environment-aware text-to-speech (TTS) model that generates natural speech seamlessly integrated within environmental contexts by explicitly modeling cross-modal interactions. Our model builds on a multimodal diffusion transformer and fuses transcript-aligned speech latent with text-conditioned environmental context via joint attention. To enhance semantic consistency, we introduce a domain-specific representation alignment objective tailored to environment-aware TTS, leveraging complementary self-supervised representations from speech and audio encoders. Experimental results show that ImmersiveTTS achieves higher naturalness, intelligibility, and audio fidelity than existing approaches across objective metrics and human listening tests.

Training automated pronunciation assessment often relies on labeled learner errors or non-native corpora that are costly to collect. We propose a lightweight framework trained only on native speech resources, operating unsupervised or lightly calibrated with a small set of scored utterances. At inference, learner speech is discretized with an SSL encoder and a K-means codebook. A token language model trained on native sequences computes surprisal where higher surprisal indicates phonotactic deviation. We add a transcript-guided Text2DUnit--DTW module that predicts native token sequences from reference text and aligns them to acoustic tokens to derive error-sensitive features. Surprisal and alignment features are fused via simple regression. On SpeechOcean762, PCC improves from 0.60 to 0.66 with transcript guidance, near supervised baselines. Cross-dataset evaluation on L2-ARCTIC shows consistent gains.

We present AdVerb, a novel audio-visual dereverberation framework that uses visual cues in addition to the reverberant sound to estimate clean audio. Although audio-only dereverberation is a well-studied problem, our approach incorporates the complementary visual modality to perform audio dereverberation. Given an image of the environment where the reverberated sound signal has been recorded, AdVerb employs a novel geometry-aware cross-modal transformer architecture that captures scene geometry and audio-visual cross-modal relationship to generate a complex ideal ratio mask, which, when applied to the reverberant audio predicts the clean sound. The effectiveness of our method is demonstrated through extensive quantitative and qualitative evaluations. Our approach significantly outperforms traditional audio-only and audio-visual baselines on three downstream tasks: speech enhancement, speech recognition, and speaker verification, with relative improvements in the range of 18% - 82% on the LibriSpeech test-clean set. We also achieve highly satisfactory RT60 error scores on the AVSpeech dataset.

Despite the potential of diffusion models in speech enhancement, their deployment in Acoustic Echo Cancellation (AEC) has been restricted. In this paper, we propose DI-AEC, pioneering a diffusion-based stochastic regeneration approach dedicated to AEC. Further, we propose FADI-AEC, fast score-based diffusion AEC framework to save computational demands, making it favorable for edge devices. It stands out by running the score model once per frame, achieving a significant surge in processing efficiency. Apart from that, we introduce a novel noise generation technique where far-end signals are utilized, incorporating both far-end and near-end signals to refine the score model's accuracy. We test our proposed method on the ICASSP2023 Microsoft deep echo cancellation challenge evaluation dataset, where our method outperforms some of the end-to-end methods and other diffusion based echo cancellation methods.

We present a novel approach for synthesizing 3D facial motions from audio sequences using key motion embeddings. Despite recent advancements in data-driven techniques, accurately mapping between audio signals and 3D facial meshes remains challenging. Direct regression of the entire sequence often leads to over-smoothed results due to the ill-posed nature of the problem. To this end, we propose a progressive learning mechanism that generates 3D facial animations by introducing key motion capture to decrease cross-modal mapping uncertainty and learning complexity. Concretely, our method integrates linguistic and data-driven priors through two modules: the linguistic-based key motion acquisition and the cross-modal motion completion. The former identifies key motions and learns the associated 3D facial expressions, ensuring accurate lip-speech synchronization. The latter extends key motions into a full sequence of 3D talking faces guided by audio features, improving temporal coherence and audio-visual consistency. Extensive experimental comparisons against existing state-of-the-art methods demonstrate the superiority of our approach in generating more vivid and consistent talking face animations. Consistent enhancements in results through the integration of our proposed learning scheme with existing methods underscore the efficacy of our approach. Our code and weights will be at the project website: https://github.com/ffxzh/KMTalk.

We introduce RASMALAI, a large-scale speech dataset with rich text descriptions, designed to advance controllable and expressive text-to-speech (TTS) synthesis for 23 Indian languages and English. It comprises 13,000 hours of speech and 24 million text-description annotations with fine-grained attributes like speaker identity, accent, emotion, style, and background conditions. Using RASMALAI, we develop IndicParlerTTS, the first open-source, text-description-guided TTS for Indian languages. Systematic evaluation demonstrates its ability to generate high-quality speech for named speakers, reliably follow text descriptions and accurately synthesize specified attributes. Additionally, it effectively transfers expressive characteristics both within and across languages. IndicParlerTTS consistently achieves strong performance across these evaluations, setting a new standard for controllable multilingual expressive speech synthesis in Indian languages.

In this work, we present TalkCuts, a large-scale dataset designed to facilitate the study of multi-shot human speech video generation. Unlike existing datasets that focus on single-shot, static viewpoints, TalkCuts offers 164k clips totaling over 500 hours of high-quality human speech videos with diverse camera shots, including close-up, half-body, and full-body views. The dataset includes detailed textual descriptions, 2D keypoints and 3D SMPL-X motion annotations, covering over 10k identities, enabling multimodal learning and evaluation. As a first attempt to showcase the value of the dataset, we present Orator, an LLM-guided multi-modal generation framework as a simple baseline, where the language model functions as a multi-faceted director, orchestrating detailed specifications for camera transitions, speaker gesticulations, and vocal modulation. This architecture enables the synthesis of coherent long-form videos through our integrated multi-modal video generation module. Extensive experiments in both pose-guided and audio-driven settings show that training on TalkCuts significantly enhances the cinematographic coherence and visual appeal of generated multi-shot speech videos. We believe TalkCuts provides a strong foundation for future work in controllable, multi-shot speech video generation and broader multimodal learning.

Hate speech detection on social media faces challenges in both accuracy and explainability, especially for underexplored Indic languages. We propose a novel explainability-guided training framework, X-MuTeST (eXplainable Multilingual haTe Speech deTection), for hate speech detection that combines high-level semantic reasoning from large language models (LLMs) with traditional attention-enhancing techniques. We extend this research to Hindi and Telugu alongside English by providing benchmark human-annotated rationales for each word to justify the assigned class label. The X-MuTeST explainability method computes the difference between the prediction probabilities of the original text and those of unigrams, bigrams, and trigrams. Final explanations are computed as the union between LLM explanations and X-MuTeST explanations. We show that leveraging human rationales during training enhances both classification performance and explainability. Moreover, combining human rationales with our explainability method to refine the model attention yields further improvements. We evaluate explainability using Plausibility metrics such as Token-F1 and IOU-F1 and Faithfulness metrics such as Comprehensiveness and Sufficiency. By focusing on under-resourced languages, our work advances hate speech detection across diverse linguistic contexts. Our dataset includes token-level rationale annotations for 6,004 Hindi, 4,492 Telugu, and 6,334 English samples. Data and code are available on https://github.com/ziarehman30/X-MuTeST

Video dubbing is a cornerstone of multimedia content creation, aiming to synthesize synchronized acoustic sequences for visual streams. While Text-to-Speech (TTS) and Text-to-Audio (TTA) generation have each achieved remarkable progress, existing dubbing systems remain confined to isolated speech synthesis without incorporating sound effects and ambient audio, forcing practitioners to rely on fragmented workflows and laborious manual post-mixing. To address this limitation, we present HoliDubber, a holistic video dubbing framework that moves beyond speech-only generation by enabling the joint synthesis of speech and sound effects from a single text prompt. Specifically, HoliDubber adopts a patch-based autoregressive diffusion transformer architecture, where a causal language model autoregressively models aggregated patch embeddings to capture global temporal structure, and a Diffusion Transformer decoder generates high-fidelity continuous tokens within each patch, following a divide-and-conquer strategy. To achieve cross-modal alignment, visual features are encoded into patch-level representations and fused with audio patches via cross-attention, enabling the model to ground speech generation in the speaker's visual articulation dynamics. In addition, we introduce HoliDub-Bench, a benchmark curated from established datasets with synchronized video-text-audio triplets designed for holistic dubbing evaluation. Extensive experiments demonstrate that HoliDubber significantly outperforms existing methods across multiple benchmarks in speech quality, synchronization, and speaker similarity. Furthermore, results on HoliDub-Bench validate the effectiveness of joint speech-and-sound generation, establishing a new paradigm for holistic video dubbing in complex acoustic scenes. The demo page of the project is https://holidubber.github.io

Target Speaker Automatic Speech Recognition (TS-ASR) aims to transcribe the speech of a specified target speaker from multi-speaker mixtures in cocktail party scenarios. Recent advancement of Large Audio-Language Models (LALMs) has already brought some new insights to TS-ASR. However, significant room for optimization remains for the TS-ASR task within the LALMs architecture. While Chain of Thoughts (CoT) and Reinforcement Learning (RL) have proven effective in certain speech tasks, TS-ASR, which requires the model to deeply comprehend speech signals, differentiate various speakers, and handle overlapping utterances is particularly well-suited to a reasoning-guided approach. Therefore, we propose a novel framework that incorporates CoT and RL training into TS-ASR for performance improvement. A novel CoT dataset of TS-ASR is constructed, and the TS-ASR model is first trained on regular data and then fine-tuned on CoT data. Finally, the model is further trained with RL using selected data to enhance generalized reasoning capabilities. Experiment results show a significant improvement of TS-ASR performance with CoT and RL training, which demonstrates the effectiveness of the proposed CoT and RL training methods adapted for the TS-ASR task.

Feature distillation is usually expected to add to logit distillation. We find that this expectation does not hold for two common feature losses in compact multilingual automatic speech recognition (ASR), and analyze why through their gradients. The gradient of a projection-based loss is orthogonal to the task gradient, while that of Centered Kernel Alignment (CKA), a similarity-based loss, is weakly aligned with it. CKA lowers the word error rate (WER) on non-English languages, but at the expense of data-rich English. Three loss-level remedies (one trained, two probed) all fall short, and the CKA gradient overlaps with the subspace English uses in the shared encoder by 97.5% to 99.5% (about 81% by chance). Guided by this diagnosis, we propose CAFE (Confined-Adapter Feature distillation), which confines the CKA gradient to adapters that English bypasses, addressing the trade-off in the parameters rather than in the loss. On the Open ASR Leaderboard, CAFE lowers the average WER from 8.55% to 8.02%, beats an adapter-only capacity control at 8.51%, and makes English more stable across checkpoints. At 254M parameters, it trails the 0.6B teacher by 2 WER points and is competitive with much larger models across 48 test sets in 24 languages. We use 564k hours of training data and state-of-the-art models to keep the findings relevant to practice.

Gestures are non-verbal but important behaviors accompanying people's speech. While previous methods are able to generate speech rhythm-synchronized gestures, the semantic context of the speech is generally lacking in the gesticulations. Although semantic gestures do not occur very regularly in human speech, they are indeed the key for the audience to understand the speech context in a more immersive environment. Hence, we introduce LivelySpeaker, a framework that realizes semantics-aware co-speech gesture generation and offers several control handles. In particular, our method decouples the task into two stages: script-based gesture generation and audio-guided rhythm refinement. Specifically, the script-based gesture generation leverages the pre-trained CLIP text embeddings as the guidance for generating gestures that are highly semantically aligned with the script. Then, we devise a simple but effective diffusion-based gesture generation backbone simply using pure MLPs, that is conditioned on only audio signals and learns to gesticulate with realistic motions. We utilize such powerful prior to rhyme the script-guided gestures with the audio signals, notably in a zero-shot setting. Our novel two-stage generation framework also enables several applications, such as changing the gesticulation style, editing the co-speech gestures via textual prompting, and controlling the semantic awareness and rhythm alignment with guided diffusion. Extensive experiments demonstrate the advantages of the proposed framework over competing methods. In addition, our core diffusion-based generative model also achieves state-of-the-art performance on two benchmarks. The code and model will be released to facilitate future research.

This paper introduces a novel voice conversion (VC) model, guided by text instructions such as "articulate slowly with a deep tone" or "speak in a cheerful boyish voice". Unlike traditional methods that rely on reference utterances to determine the attributes of the converted speech, our model adds versatility and specificity to voice conversion. The proposed VC model is a neural codec language model which processes a sequence of discrete codes, resulting in the code sequence of converted speech. It utilizes text instructions as style prompts to modify the prosody and emotional information of the given speech. In contrast to previous approaches, which often rely on employing separate encoders like prosody and content encoders to handle different aspects of the source speech, our model handles various information of speech in an end-to-end manner. Experiments have demonstrated the impressive capabilities of our model in comprehending instructions and delivering reasonable results.

Classifier-free guidance is a key component for enhancing the performance of conditional generative models across diverse tasks. While it has previously demonstrated remarkable improvements for the sample quality, it has only been exclusively employed for diffusion models. In this paper, we integrate classifier-free guidance into Flow Matching (FM) models, an alternative simulation-free approach that trains Continuous Normalizing Flows (CNFs) based on regressing vector fields. We explore the usage of Guided Flows for a variety of downstream applications. We show that Guided Flows significantly improves the sample quality in conditional image generation and zero-shot text-to-speech synthesis, boasting state-of-the-art performance. Notably, we are the first to apply flow models for plan generation in the offline reinforcement learning setting, showcasing a 10x speedup in computation compared to diffusion models while maintaining comparable performance.

Talking face generation, also known as speech-to-lip generation, reconstructs facial motions concerning lips given coherent speech input. The previous studies revealed the importance of lip-speech synchronization and visual quality. Despite much progress, they hardly focus on the content of lip movements i.e., the visual intelligibility of the spoken words, which is an important aspect of generation quality. To address the problem, we propose using a lip-reading expert to improve the intelligibility of the generated lip regions by penalizing the incorrect generation results. Moreover, to compensate for data scarcity, we train the lip-reading expert in an audio-visual self-supervised manner. With a lip-reading expert, we propose a novel contrastive learning to enhance lip-speech synchronization, and a transformer to encode audio synchronically with video, while considering global temporal dependency of audio. For evaluation, we propose a new strategy with two different lip-reading experts to measure intelligibility of the generated videos. Rigorous experiments show that our proposal is superior to other State-of-the-art (SOTA) methods, such as Wav2Lip, in reading intelligibility i.e., over 38% Word Error Rate (WER) on LRS2 dataset and 27.8% accuracy on LRW dataset. We also achieve the SOTA performance in lip-speech synchronization and comparable performances in visual quality.

Recent work reports gains in neural text-to-speech (TTS) with Group Relative Policy Optimization (GRPO). However, in the absence of a verifiable reward for prosody, GRPO trained on transcription-oriented signals (CER/NLL) lowers error rates yet collapses prosody into monotone, unnatural speech; adding speaker-similarity further destabilizes training and degrades CER. We address this with an iterative Direct Preference Optimization (DPO) scheme that uses only a few hundred human-labeled preference pairs per round to directly optimize prosodic naturalness while regularizing to the current model. On KoCC-TTS, a curated dataset of authentic Korean call center interactions capturing task-oriented dialogues, our method attains the highest human preference (ELO) with competitive CER, outperforming GRPO and strong commercial baselines. These results suggest that when prosody cannot be rewarded automatically, human preference optimization offers a practical and data-efficient path to natural and robust TTS. The demo page is available at https://tts.ch.dev

Developing high-quality text-to-speech (TTS) systems for low-resource languages is challenging due to the scarcity of paired text and speech data. In contrast, automatic speech recognition (ASR) models for such languages are often more accessible, owing to large-scale multilingual pre-training efforts. We propose a framework based on Group Relative Policy Optimization (GRPO) to adapt an autoregressive, multilingual TTS model to new languages. Our method first establishes a language-agnostic foundation for TTS synthesis by training a multilingual baseline with International Phonetic Alphabet (IPA) tokens. Next, we fine-tune this model on limited paired data of the new languages to capture the target language's prosodic features. Finally, we apply GRPO to optimize the model using only unpaired text and speaker prompts, guided by a multi-objective reward from pretrained ASR, speaker verification, and audio quality estimation models. Experiments demonstrate that this pipeline produces intelligible and speaker-consistent speech in low-resource languages, substantially outperforming fine-tuning alone. Furthermore, our GRPO-based framework also improves TTS performance in high-resource languages, surpassing offline alignment methods such as Direct Preference Optimization (DPO) yielding superior intelligibility, speaker similarity, and audio quality.

Imperceptible text-based speech editing modifies spoken content through transcript manipulation while preserving acoustic continuity. Prior acoustic-space approaches suffer from content-style entanglement, causing unstable generation and boundary artifacts. We introduce a framework guided by the principle of "Edit Content, Preserve Acoustics". Editing is conducted in a stable semantic space, while acoustic realization is handled by a Flow Matching decoder. To ensure perceptual consistency, we propose Self-Consistency Rewards Group Relative Policy Optimization, which leverages a pre-trained Text-to-Speech model as an implicit critic, together with intelligibility and duration constraints. Experiments demonstrate consistent improvements over state-of-the-art autoregressive and non-autoregressive baselines in intelligibility, robustness, and perceptual quality.

Automated transcription of parliamentary proceedings faces significant hurdles due to demographic bias, dialectal variation, and technical artifacts such as utterance truncation during segmentation. This paper introduces the ROManian PARliamentary Speech Corpus (ROMPAR) dataset, a 17.80-hour corpus of Romanian and Moldavian parliamentary speech, featuring double-annotated ground truth and explicit labels for reconstructed word fragments. To build a robust ASR system, we propose a multi-task adversarial training framework that enforces demographic invariance across age, gender, and dialect. We address the inherent instability of adversarial objectives in generative architectures by introducing an exponential decay mechanism for the adversarial coefficients. Furthermore, we implement an LLM-guided decoding strategy with position-dependent weighting to facilitate morphological completion of truncated terminal words. Our results demonstrate that the proposed framework significantly reduces WER and achieves an F1-score of 96.6% in morphological reconstruction.

Non-parallel many-to-many voice conversion remains an interesting but challenging speech processing task. Recently, AutoVC, a conditional autoencoder based method, achieved excellent conversion results by disentangling the speaker identity and the speech content using information-constraining bottlenecks. However, due to the pure autoencoder training method, it is difficult to evaluate the separation effect of content and speaker identity. In this paper, a novel voice conversion framework, named boldsymbol Text boldsymbol Guided boldsymbol AutoVC(TGAVC), is proposed to more effectively separate content and timbre from speech, where an expected content embedding produced based on the text transcriptions is designed to guide the extraction of voice content. In addition, the adversarial training is applied to eliminate the speaker identity information in the estimated content embedding extracted from speech. Under the guidance of the expected content embedding and the adversarial training, the content encoder is trained to extract speaker-independent content embedding from speech. Experiments on AIShell-3 dataset show that the proposed model outperforms AutoVC in terms of naturalness and similarity of converted speech.

Recent portrait animation methods have made significant strides in generating realistic lip synchronization. However, they often lack explicit control over head movements and facial expressions, and cannot produce videos from multiple viewpoints, resulting in less controllable and expressive animations. Moreover, text-guided portrait animation remains underexplored, despite its user-friendly nature. We present a novel two-stage text-guided framework, MVPortrait (Multi-view Vivid Portrait), to generate expressive multi-view portrait animations that faithfully capture the described motion and emotion. MVPortrait is the first to introduce FLAME as an intermediate representation, effectively embedding facial movements, expressions, and view transformations within its parameter space. In the first stage, we separately train the FLAME motion and emotion diffusion models based on text input. In the second stage, we train a multi-view video generation model conditioned on a reference portrait image and multi-view FLAME rendering sequences from the first stage. Experimental results exhibit that MVPortrait outperforms existing methods in terms of motion and emotion control, as well as view consistency. Furthermore, by leveraging FLAME as a bridge, MVPortrait becomes the first controllable portrait animation framework that is compatible with text, speech, and video as driving signals.

This paper describes our NPU-ASLP system for the Audio-Visual Diarization and Recognition (AVDR) task in the Multi-modal Information based Speech Processing (MISP) 2022 Challenge. Specifically, the weighted prediction error (WPE) and guided source separation (GSS) techniques are used to reduce reverberation and generate clean signals for each single speaker first. Then, we explore the effectiveness of Branchformer and E-Branchformer based ASR systems. To better make use of the visual modality, a cross-attention based multi-modal fusion module is proposed, which explicitly learns the contextual relationship between different modalities. Experiments show that our system achieves a concatenated minimum-permutation character error rate (cpCER) of 28.13\% and 31.21\% on the Dev and Eval set, and obtains second place in the challenge.

Diffusion-based text-to-speech (TTS) systems have made remarkable progress in zero-shot speech synthesis, yet optimizing all components for perceptual metrics remains challenging. Prior work with DMOSpeech demonstrated direct metric optimization for speech generation components, but duration prediction remained unoptimized. This paper presents DMOSpeech 2, which extends metric optimization to the duration predictor through a reinforcement learning approach. The proposed system implements a novel duration policy framework using group relative preference optimization (GRPO) with speaker similarity and word error rate as reward signals. By optimizing this previously unoptimized component, DMOSpeech 2 creates a more complete metric-optimized synthesis pipeline. Additionally, this paper introduces teacher-guided sampling, a hybrid approach leveraging a teacher model for initial denoising steps before transitioning to the student model, significantly improving output diversity while maintaining efficiency. Comprehensive evaluations demonstrate superior performance across all metrics compared to previous systems, while reducing sampling steps by half without quality degradation. These advances represent a significant step toward speech synthesis systems with metric optimization across multiple components. The audio samples, code and pre-trained models are available at https://dmospeech2.github.io/.

Speech Relation Extraction (SpeechRE) aims to extract relation triplets directly from speech. However, existing benchmark datasets rely heavily on synthetic data, lacking sufficient quantity and diversity of real human speech. Moreover, existing models also suffer from rigid single-order generation templates and weak semantic alignment, substantially limiting their performance. To address these challenges, we introduce CommonVoice-SpeechRE, a large-scale dataset comprising nearly 20,000 real-human speech samples from diverse speakers, establishing a new benchmark for SpeechRE research. Furthermore, we propose the Relation Prompt-Guided Multi-Order Generative Ensemble (RPG-MoGe), a novel framework that features: (1) a multi-order triplet generation ensemble strategy, leveraging data diversity through diverse element orders during both training and inference, and (2) CNN-based latent relation prediction heads that generate explicit relation prompts to guide cross-modal alignment and accurate triplet generation. Experiments show our approach outperforms state-of-the-art methods, providing both a benchmark dataset and an effective solution for real-world SpeechRE. The source code and dataset are publicly available at https://github.com/NingJinzhong/SpeechRE_RPG_MoGe.

Existing Visual Speech Recognition (VSR) systems commonly rely on left-to-right autoregressive decoding, which can force premature decisions on visually ambiguous tokens before sufficient context is available. We propose DLLM-VSR, to the best of our knowledge, the first Diffusion Large Language Model (DLLM)-based VSR framework, formulating transcription as iterative masked denoising with flexible-order decoding. With confidence-based unmasking, DLLM-VSR commits high-confidence positions early and uses the committed tokens as bidirectional context to refine ambiguous ones. To adapt DLLMs to VSR, we introduce a two-stage masked-denoising training strategy that separates visual-to-text content alignment from length modeling. We further observe a performance gap with oracle-length decoding, which assumes access to the true transcript length, indicating that reducing target-length uncertainty can improve DLLM-based VSR. To reduce this gap, we develop length-guided candidate decoding, which uses video duration to construct plausible transcript-length hypotheses, decodes under multiple hypotheses, and reranks candidates using length plausibility and decoding confidence. The proposed method achieves a state-of-the-art WER of 19.5\% on LRS3 using only its labeled training data.

Real-time long-form digital-human generation relies on causal models to extend audio-visual content while preserving subject appearance and audio-video synchronization across successive segments. A bounded cache retains local motion and phonetic context but discards older evidence, whereas attending to the complete generated history is computationally expensive and can propagate accumulated errors. We present \method, an anchor-guided persistent-memory framework for few-step joint audio-video generation. The framework preserves an immutable visual anchor, compresses completed video and audio blocks into fixed-capacity dynamic states, and retrieves those states through modality-specific residual attention without extending the active cache. A reference-aware modulation method additionally conditions video features on dynamic and anchor appearance statistics. Anchor-preserving causal-context distillation varies rollout horizon, prefix provenance, and cache-history reliability while keeping the immutable visual anchor unperturbed. By separating persistent memory from stage-local denoising dependencies, \method further admits stage-parallel execution across blocks, accelerating autoregressive inference without pipeline-specific retraining. We evaluate long-form video continuations with appearance, temporal, synchronization, facial, and speech diagnostics. Results show that \method preserves stable appearance across prompt-conditioned segments and strong audio-visual synchronization under autoregressive generation. Our project page is https://taoliveaigc.github.io/TaoMate.

Self-supervised learning (SSL) countermeasures (CMs) have shown strong performance in recent years. However, they often show degraded performance while facing unseen spoofing attacks and mismatched conditions. This study examines the Voxtral audio-language model (ALM) framework for spoofing detection, as a step toward combining CM capabilities within the ALM framework. We analyze how Voxtral captures spoofing cues through audio-text processing and propose an instruction-guided approach that uses label-sequence likelihoods to evaluate bonafide and spoofed speech. Experiments on the ASVspoof databases show that without task-specific adaptation, the LLM layers emphasize semantic representations, reducing the separability of spoof-discriminative acoustic cues compared to the Whisper-based audio encoder. Consequently, spoofing-related information becomes less separable after language-model processing. We also applied lightweight adaptation using weight-decomposed low-rank adaptation (DoRA) to the Voxtral model and propose the Spooftral model, achieving an equal error rate (EER) of 4.25% on the ASVspoof5 evaluation set.

We introduce a novel method for joint expression and audio-guided talking face generation. Recent approaches either struggle to preserve the speaker identity or fail to produce faithful facial expressions. To address these challenges, we propose a NeRF-based network. Since we train our network on monocular videos without any ground truth, it is essential to learn disentangled representations for audio and expression. We first learn audio features in a self-supervised manner, given utterances from multiple subjects. By incorporating a contrastive learning technique, we ensure that the learned audio features are aligned to the lip motion and disentangled from the muscle motion of the rest of the face. We then devise a transformer-based architecture that learns expression features, capturing long-range facial expressions and disentangling them from the speech-specific mouth movements. Through quantitative and qualitative evaluation, we demonstrate that our method can synthesize high-fidelity talking face videos, achieving state-of-the-art facial expression transfer along with lip synchronization to unseen audio.

In this paper, we propose a novel audio-driven talking head method capable of simultaneously generating highly expressive facial expressions and hand gestures. Unlike existing methods that focus on generating full-body or half-body poses, we investigate the challenges of co-speech gesture generation and identify the weak correspondence between audio features and full-body gestures as a key limitation. To address this, we redefine the task as a two-stage process. In the first stage, we generate hand poses directly from audio input, leveraging the strong correlation between audio signals and hand movements. In the second stage, we employ a diffusion model to synthesize video frames, incorporating the hand poses generated in the first stage to produce realistic facial expressions and body movements. Our experimental results demonstrate that the proposed method outperforms state-of-the-art approaches, such as CyberHost and Vlogger, in terms of both visual quality and synchronization accuracy. This work provides a new perspective on audio-driven gesture generation and a robust framework for creating expressive and natural talking head animations.

Automatic speech recognition (ASR) systems based on large language models (LLMs) achieve superior performance by leveraging pretrained LLMs as decoders, but their token-by-token generation mechanism leads to inference latency that grows linearly with sequence length. Meanwhile, discrete diffusion large language models (dLLMs) offer a promising alternative, enabling high-quality parallel sequence generation with pretrained decoders. However, directly applying native text-oriented dLLMs to ASR leads to a fundamental mismatch between open-ended text generation and the acoustically conditioned transcription paradigm required by ASR. As a result, it introduces unnecessary difficulty and computational redundancy, such as denoising from pure noise, inflexible generation lengths, and fixed denoising steps. We propose dLLM-ASR, an efficient dLLM-based ASR framework that formulates dLLM's decoding as a prior-guided and adaptive denoising process. It leverages an ASR prior to initialize the denoising process and provide an anchor for sequence length. Building upon this prior, length-adaptive pruning dynamically removes redundant tokens, while confidence-based denoising allows converged tokens to exit the denoising loop early, enabling token-level adaptive computation. Experiments demonstrate that dLLM-ASR achieves recognition accuracy comparable to autoregressive LLM-based ASR systems and delivers a 4.44times inference speedup, establishing a practical and efficient paradigm for ASR.

Universal Speech Enhancement (USE) aims to restore speech quality under diverse degradation conditions while preserving signal fidelity. Despite recent progress, key challenges in training target selection, the distortion--perception tradeoff, and data curation remain unresolved. In this work, we systematically address these three overlooked problems. First, we revisit the conventional practice of using early-reflected speech as the dereverberation target and show that it can degrade perceptual quality and downstream ASR performance. We instead demonstrate that time-shifted anechoic clean speech provides a superior learning target. Second, guided by the distortion--perception tradeoff theory, we propose a simple two-stage framework that achieves minimal distortion under a given level of perceptual quality. Third, we analyze the trade-off between training data scale and quality for USE, revealing that training on large uncurated corpora imposes a performance ceiling, as models struggle to remove subtle artifacts. Our method achieves state-of-the-art performance on the URGENT 2025 non-blind test set and exhibits strong language-agnostic generalization, making it effective for improving TTS training data. Code and models will be released upon acceptance.

Fine-grained local timing control is still absent from modern text-to-speech systems: existing approaches typically provide only utterance-level duration or global speaking-rate control, while precise token-level timing manipulation remains unavailable. To the best of our knowledge, MAGIC-TTS is the first TTS model with explicit local timing control over token-level content duration and pause. MAGIC-TTS is enabled by explicit token-level duration conditioning, carefully prepared high-confidence duration supervision, and training mechanisms that correct zero-value bias and make the model robust to missing local controls. On our timing-control benchmark, MAGIC-TTS substantially improves token-level duration and pause following over spontaneous synthesis. Even when no timing control is provided, MAGIC-TTS maintains natural high-quality synthesis. We further evaluate practical local editing with a scenario-based benchmark covering navigation guidance, guided reading, and accessibility-oriented code reading. In this setting, MAGIC-TTS realizes a reproducible uniform-timing baseline and then moves the edited regions toward the requested local targets with low mean bias. These results show that explicit fine-grained controllability can be implemented effectively in a high-quality TTS system and can support realistic local timing-editing applications.

Recent advances in voice cloning and text-to-speech synthesis have made partial speech manipulation - where an adversary replaces a few words within an utterance to alter its meaning while preserving the speaker's identity - an increasingly realistic threat. Existing audio deepfake detection benchmarks focus on utterance-level binary classification or single-region tampering, leaving a critical gap in detecting and localizing multiple inpainted segments whose count is unknown a priori. We address this gap with three contributions. First, we introduce MIST (Multiregion Inpainting Speech Tampering), a large-scale multilingual dataset spanning 6 languages with 1-3 independently inpainted word-level segments per utterance, generated via LLM-guided semantic replacement and neural voice cloning, with fake content constituting only 2-7% of each utterance. Second, we propose ISA (Iterative Segment Analysis), a backbone-agnostic framework that performs coarse-to-fine sliding-window classification with gap-tolerant region proposal and boundary refinement to recover all tampered regions without prior knowledge of their count. Third, we define SF1@tau, a segment-level F1 metric based on temporal IoU matching that jointly evaluates region count accuracy and localization precision. Zero-shot evaluation reveals that partial inpainting at word granularity remains unsolved by existing deepfake detectors: utterance-level classifiers trained on fully synthesized speech assign near zero fake probability to MIST utterances where only 2-7% of content is manipulated. ISA consistently outperforms non-iterative baselines in this challenging setting, and the dataset, code, and evaluation toolkit are publicly released.

Building competitive automatic speech recognition (ASR) models usually requires large-scale au- dio supervision, which makes reproduction and specialization expensive. We study Ark-ASR, a 0.6B- parameter audio-conditioned language model trained with 100k hours of speech, and examine whether a strong Qwen-ASR teacher can transfer additional recognition capability through on-policy distillation. Across Mandarin and English ASR benchmarks, the proposed training recipe consistently improves over supervised fine-tuning alone and outperforms the same-scale Qwen3-ASR-0.6B baseline on four of five evaluation sets. This is achieved with only 100k hours of speech, compared with the 20M hours of super- vised audio reported for the Qwen3-Omni AuT encoder. The larger Qwen3-ASR-1.7B remains stronger, but the results show that teacher-guided on-policy training can substantially close the gap for compact ASR models under a much smaller audio budget. A support-overlap diagnostic further suggests that the teacher-data stage improves local student-teacher compatibility, matching recent analyses of when on-policy distillation is effective.

Motion, speech, and sound effects are fundamental elements of human-centric videos, yet their heterogeneous temporal characteristics make joint generation highly challenging. Existing audio-video generation models often fail to maintain consistent alignment across these modalities, leading to noticeable mismatches between motion, speech, and environmental sounds. We present Unison, a unified framework that explicitly promotes coherence across the motion, speech, and sound modalities. Within the audio stream, Unison employs a semantic-guided harmonization strategy that decouples the generation of speech and sound-effect components. Leveraging bidirectional audio cross-attention and semantic-conditioned gating for semantic-driven adaptive recomposition, this approach effectively mitigates speech dominance and enhances acoustic clarity. For audio-motion synchronization, we propose a bidirectional cross-modal forcing strategy where the cleaner modality guides the noisier one through decoupled denoising schedules, reinforced by a progressive stabilization strategy. Extensive experiments demonstrate that Unison achieves state-of-the-art performance in both audio perceptual quality and cross-modal synchronization, highlighting the importance of explicit multimodal harmonization in human-centric video generation.

While instruction-based video editing has seen significant progress, joint audio-visual editing remains constrained by the absence of dedicated datasets and benchmarks. To bridge this gap, we present JAVEdit-100k, the first large-scale, high-quality dataset tailored for instruction-guided joint audio-visual editing. Focusing on human-centric videos, JAVEdit-100k comprises approximately 100K editing triplets spanning five distinct categories, including subject editing and speech editing. This dataset is rigorously constructed via four meticulously designed generation pipelines, seamlessly paired with an agent-in-the-loop quality control mechanism. Furthermore, to address the lack of standardized evaluation within the field, we introduce JAVEditBench, a comprehensive benchmark featuring curated source videos and human-aligned instructions across all editing categories. Finally, we propose JAVEdit, a pioneering baseline model for instruction-guided joint audio-visual editing. Experiments show that \model\ outperforms all baselines on five of six evaluation metrics.

We introduce AuK, an open-source foundational model that unifies speech generation and editing through a common interface of natural-language instructions and audio context. To support this broad capability set, we construct approximately 3.03 billion instruction--audio instances and 1.95 million hours of effective supervision across five task families: speech generation, content editing, enhancement and separation, paralinguistic editing, and acoustic editing. AuK combines a multimodal large language model for semantic conditioning, an VAE jointly trained on speech, general audio, and music for acoustic conditioning, and a hybrid rectified-flow Transformer that performs dual-stream MMDiT blocks followed by unified single-stream DiT blocks for generation. Training begins with generation-only warm-up and proceeds to joint generation--editing pre-training. We then apply complementary post-training strategies: human-feedback preference optimization for open-ended editing and reward-based reinforcement learning for speech generation. To reduce inference cost, we further distill the model with consistency initialization and task-routed Decoupled DMD. The resulting AuK-Flash performs 4-step inference without classifier-free guidance and achieves a 4.5 wall-clock speedup over the full model under matched conditions. Experiments demonstrate leading performance on zero-shot and instruction-controlled speech generation and general instruction-guided editing, while remaining competitive on signal-level restoration tasks. We release both the source code and model weights to support reproducibility and further research.

Multilingual speech recognition for both monolingual and code-switching speech is a challenging task. Recently, based on the Mixture of Experts (MoE), many works have made good progress in multilingual and code-switching ASR, but present huge computational complexity with the increase of supported languages. In this work, we propose a computation-efficient network named Language-Routing Mixture of Experts (LR-MoE) for multilingual and code-switching ASR. LR-MoE extracts language-specific representations through the Mixture of Language Experts (MLE), which is guided to learn by a frame-wise language routing mechanism. The weight-shared frame-level language identification (LID) network is jointly trained as the shared pre-router of each MoE layer. Experiments show that the proposed method significantly improves multilingual and code-switching speech recognition performances over baseline with comparable computational efficiency.

In daily life, we encounter a variety of sounds, both desirable and undesirable, with limited control over their presence and volume. Our work introduces "Listen, Chat, and Edit" (LCE), a novel multimodal sound mixture editor that modifies each sound source in a mixture based on user-provided text instructions. LCE distinguishes itself with a user-friendly chat interface and its unique ability to edit multiple sound sources simultaneously within a mixture, without needing to separate them. Users input open-vocabulary text prompts, which are interpreted by a large language model to create a semantic filter for editing the sound mixture. The system then decomposes the mixture into its components, applies the semantic filter, and reassembles it into the desired output. We developed a 160-hour dataset with over 100k mixtures, including speech and various audio sources, along with text prompts for diverse editing tasks like extraction, removal, and volume control. Our experiments demonstrate significant improvements in signal quality across all editing tasks and robust performance in zero-shot scenarios with varying numbers and types of sound sources.

While autoregressive speech token generation models produce speech with remarkable variety and naturalness, their inherent lack of controllability often results in issues such as hallucinations and undesired vocalizations that do not conform to conditioning inputs. We introduce Koel-TTS, a suite of enhanced encoder-decoder Transformer TTS models that address these challenges by incorporating preference alignment techniques guided by automatic speech recognition and speaker verification models. Additionally, we incorporate classifier-free guidance to further improve synthesis adherence to the transcript and reference speaker audio. Our experiments demonstrate that these optimizations significantly enhance target speaker similarity, intelligibility, and naturalness of synthesized speech. Notably, Koel-TTS directly maps text and context audio to acoustic tokens, and on the aforementioned metrics, outperforms state-of-the-art TTS models, despite being trained on a significantly smaller dataset. Audio samples and demos are available on our website.

This paper introduces Swivuriso, a 3000-hour multilingual speech dataset developed as part of the African Next Voices project, to support the development and benchmarking of automatic speech recognition (ASR) technologies in seven South African languages. Covering agriculture, healthcare, and general domain topics, Swivuriso addresses significant gaps in existing ASR datasets. We describe the design principles, ethical considerations, and data collection procedures that guided the dataset creation. We present baseline results of training/finetuning ASR models with this data and compare to other ASR datasets for the langauges concerned.

Full-duplex spoken dialogue models are trained on conversational speech in which each speaker is represented as a separate stream, but existing large-scale public speech corpora are mostly monaural, making them unsuited for SDLM training. We present DuplexChat, an open-source corpus for full-duplex spoken dialogue models, and DuplexChat-Pipe, a pipeline for constructing speaker-separated full-duplex dialogue speech from public podcast feeds. DuplexChat-Pipe filters language-specific podcast feeds, retrieves and cleans episode audio, extracts diarization-guided two-speaker dialogue clips, and applies speech separation and restoration to produce one channel per speaker. Running this pipeline yields a speaker-separated spoken dialogue corpus covering 282,634 hours of English and 132,723 hours of Japanese. Analysis results on DuplexChat show that it contains turn-taking dynamics present in human dialogues.

The language model (LM) approach based on acoustic and linguistic prompts, such as VALL-E, has achieved remarkable progress in the field of zero-shot audio generation. However, existing methods still have some limitations: 1) repetitions, transpositions, and omissions in the output synthesized speech due to limited alignment constraints between audio and phoneme tokens; 2) challenges of fine-grained control over the synthesized speech with autoregressive (AR) language model; 3) infinite silence generation due to the nature of AR-based decoding, especially under the greedy strategy. To alleviate these issues, we propose ELLA-V, a simple but efficient LM-based zero-shot text-to-speech (TTS) framework, which enables fine-grained control over synthesized audio at the phoneme level. The key to ELLA-V is interleaving sequences of acoustic and phoneme tokens, where phoneme tokens appear ahead of the corresponding acoustic tokens. The experimental findings reveal that our model outperforms VALL-E in terms of accuracy and delivers more stable results using both greedy and sampling-based decoding strategies. The code of ELLA-V will be open-sourced after cleanups. Audio samples are available at https://ereboas.github.io/ELLAV/.

Large Vision-Language Models (LVLMs) demonstrate impressive capabilities in generating detailed and coherent responses from visual inputs. However, they are prone to generate hallucinations due to an over-reliance on language priors. To address this issue, we investigate the language priors in LVLMs and make two key observations: (1) Even when predicting the tokens associated with image-related part-of-speech (POS), models increasingly rely on linguistic priors as the token sequences grow, thereby amplifying hallucinations. (2) Methods that directly calibrate LVLM's output distribution to mitigate language priors can lead to a degradation in text quality or even exacerbate hallucinations. Based on these findings, we propose a novel method, Summary-Guided Decoding (SumGD). This method naturally encourages the model to focus more on image information by reducing the text context through summaries, while controlling only the image-related POS tokens to maintain text quality. Through experiments, we demonstrate that SumGD achieves state-of-the-art performance on object hallucination benchmarks. Furthermore, in terms of the trade-off between precision and recall, SumGD achieves Pareto optimality among the existing methods. Lastly, we observe that although existing methods struggle to balance the reduction of object hallucinations with maintaining text quality, SumGD demonstrates robustness in handling this challenge.

Person-generic audio-driven face generation is a challenging task in computer vision. Previous methods have achieved remarkable progress in audio-visual synchronization, but there is still a significant gap between current results and practical applications. The challenges are two-fold: 1) Preserving unique individual traits for achieving high-precision lip synchronization. 2) Generating high-quality facial renderings in real-time performance. In this paper, we propose a novel generalized audio-driven framework RealTalk, which consists of an audio-to-expression transformer and a high-fidelity expression-to-face renderer. In the first component, we consider both identity and intra-personal variation features related to speaking lip movements. By incorporating cross-modal attention on the enriched facial priors, we can effectively align lip movements with audio, thus attaining greater precision in expression prediction. In the second component, we design a lightweight facial identity alignment (FIA) module which includes a lip-shape control structure and a face texture reference structure. This novel design allows us to generate fine details in real-time, without depending on sophisticated and inefficient feature alignment modules. Our experimental results, both quantitative and qualitative, on public datasets demonstrate the clear advantages of our method in terms of lip-speech synchronization and generation quality. Furthermore, our method is efficient and requires fewer computational resources, making it well-suited to meet the needs of practical applications.

Current TTS systems typically rely on open-loop, single-pass generation and can produce sporadic local prosodic defects, such as misplaced stress, unnatural pauses, or flattened intonation, that utterance-level metrics often fail to expose. We present LoopTTS, a judge-guided Filter-Judge-Refiner framework for recovering low-quality TTS outputs diagnosed by an AudioLLM. Given an initial utterance from a base TTS model, an AudioLLM Judge identifies salient prosodic issues and generates structured refine instructions; a Refiner, our fine-grained instruction-following TTS model, then performs guided expressive re-synthesis conditioned on the initial utterance, target text, and instruction. To train the Refiner, we construct Refiner-DB, a 42K-example AudioLLM-annotated dataset with word-level prosodic weak supervision. Human evaluation on diagnosed low-quality utterances shows that LoopTTS can detect perceptually salient errors and correct them with the Refiner, outperforming raw generated audio and practical open-loop re-generation baselines in recovery quality. The Refiner also demonstrates stronger instruction-following ability for stress and pause control in targeted prosody modification.

Text-to-speech (TTS) systems are conventionally optimized for intelligibility and speaker similarity. However, neither objective captures the prosodic detail that distinguishes human performances from synthetic speech, leaving a realism gap that larger models and higher-quality data can narrow but not close. In the absence of an objective that directly targets this gap, realism remains a persistent challenge in TTS. We present Audio8-TTS, a compact 0.6B-parameter model for zero-shot voice cloning that achieves state-of-the-art performance in intelligibility and speaker similarity, with particularly strong results in perceptual naturalness. To align the training signal with human perception rather than signal-level fidelity, we construct a quality ladder from widely used open-source TTS systems and train a realism critic to assess generated speech against matched human recordings of the same sentences. An identity-erasure objective discourages the critic from relying on generator fingerprints. We train the model in three stages: multilingual supervised fine-tuning; group relative policy optimization (GRPO) with word error rate (WER) and speaker similarity (SIM) rewards; and GRPO guided by the realism critic. The final stage improves perceived prosodic naturalness and speaker similarity while preserving WER performance.

Recent years have witnessed rapid progress in joint audio--visual generation. However, audio--visual synchronization remains a challenge, as generated speech and sound often temporally misalign with visual events. A key bottleneck is the lack of human-annotated data, reliable evaluation benchmarks, and synchronization-aware evaluation models for generated audio--visual outputs. To address these limitations, we conduct a comprehensive study spanning data, benchmarking, and reward modeling. We first introduce SyncReward-Data-80K, a dataset containing 80K professional synchronization ratings for clips from eight audio--visual generators. A taxonomy-guided construction pipeline covers four sound-event families: Speech, Onset, Instrument, and Ambient. We further construct SyncReward-Bench, a held-out benchmark of 1K clips, each independently rated by multiple professionals. Using this benchmark, we systematically evaluate specialized synchronization metrics and omni-modal models, revealing their substantial misalignment with human judgments. Finally, we develop SyncReward, a reward model that combines semantic--temporal contrastive pretraining on real clips with human supervision on generated clips. As an evaluation metric, SyncReward achieves the strongest agreement with human ratings among the evaluated methods. As a reward, it improves synchronization through inference-time best-of-N selection and reinforcement learning based post-training. We will release the data, benchmark, model, and code at https://github.com/WeChatCV/SyncReward.

Large audio-language models (LALMs) can reason about audio, yet it remains unclear whether they can perform comparative judgments between two speech signals along emotional, environmental, linguistic, prosodic, and interpersonal dimensions. We study this question in the context of speech emotion recognition (SER), where the model determines which utterance exhibits higher arousal, valence, or dominance. We introduce a reasoning-guided ordinal SER framework that conditions an LALM on paired speech inputs. The model is trained using reasoning traces generated from both semantic audio descriptions and acoustic evidence derived from GeMAPS features, enabling interpretable comparative decisions. Beyond direct supervision, we also employ direct preference optimization to encourage stronger separation for emotional differences. Experiments show that the proposed framework improves preference prediction while requiring only 5% of the training data used by conventional ordinal SER systems.

Speech tokenization enables discrete representation and facilitates speech language modeling. However, existing neural codecs capture low-level acoustic features, overlooking the semantic and contextual cues inherent to human speech. While recent efforts introduced semantic representations from self-supervised speech models or incorporated contextual representations from pre-trained language models, challenges remain in aligning and unifying the semantic and contextual representations. We introduce FuseCodec, which unifies acoustic, semantic, and contextual representations through strong cross-modal alignment and globally informed supervision. We propose three complementary techniques: (i) Latent Representation Fusion, integrating semantic and contextual features directly into the encoder latent space for robust and unified representation learning; (ii) Global Semantic-Contextual Supervision, supervising discrete tokens with globally pooled and broadcasted representations to enhance temporal consistency and cross-modal alignment; and (iii) Temporally Aligned Contextual Supervision, strengthening alignment by dynamically matching contextual and speech tokens within a local window for fine-grained token-level supervision. We further introduce FuseCodec-TTS, demonstrating our methodology's applicability to zero-shot speech synthesis. Empirically, FuseCodec achieves state-of-the-art performance in LibriSpeech, surpassing EnCodec, SpeechTokenizer, and DAC in transcription accuracy, perceptual quality, intelligibility, and speaker similarity. Results highlight the effectiveness of contextually and semantically guided tokenization for speech tokenization and downstream tasks. Code and pretrained models are available at https://github.com/mubtasimahasan/FuseCodec.

We present PFluxTTS, a hybrid text-to-speech system addressing three gaps in flow-matching TTS: the stability-naturalness trade-off, weak cross-lingual voice cloning, and limited audio quality from low-rate mel features. Our contributions are: (1) a dual-decoder design combining duration-guided and alignment-free models through inference-time vector-field fusion; (2) robust cloning using a sequence of speech-prompt embeddings in a FLUX-based decoder, preserving speaker traits across languages without prompt transcripts; and (3) a modified PeriodWave vocoder with super-resolution to 48 kHz. On cross-lingual in-the-wild data, PFluxTTS clearly outperforms F5-TTS, FishSpeech, and SparkTTS, matches ChatterBox in naturalness (MOS 4.11) while achieving 23% lower WER (6.9% vs. 9.0%), and surpasses ElevenLabs in speaker similarity (+0.32 SMOS). The system remains robust in challenging scenarios where most open-source models fail, while requiring only short reference audio and no extra training. Audio demos are available at https://braskai.github.io/pfluxtts/

Generating lifelike conversational avatars requires modeling not just isolated speakers, but the dynamic, reciprocal interaction of speaking and listening. However, modeling the listener is exceptionally challenging: direct audio-driven training fails, producing stiff, static listening motions. This failure stems from a fundamental imbalance: the speaker's motion is strongly driven by speech audio, while the listener's motion primarily follows an internal motion prior and is only loosely guided by external speech. This challenge has led most methods to focus on speak-only generation. The only prior attempt at joint generation relies on extra speaker's motion to produce the listener. This design is not end-to-end, thereby hindering the real-time applicability. To address this limitation, we present UniLS, the first end-to-end framework for generating unified speak-listen expressions, driven by only dual-track audio. Our method introduces a novel two-stage training paradigm. Stage 1 first learns the internal motion prior by training an audio-free autoregressive generator, capturing the spontaneous dynamics of natural facial motion. Stage 2 then introduces the dual-track audio, fine-tuning the generator to modulate the learned motion prior based on external speech cues. Extensive evaluations show UniLS achieves state-of-the-art speaking accuracy. More importantly, it delivers up to 44.1\% improvement in listening metrics, generating significantly more diverse and natural listening expressions. This effectively mitigates the stiffness problem and provides a practical, high-fidelity audio-driven solution for interactive digital humans.

Large Language Models (LLMs) have shown great promise in integrating diverse expert models to tackle intricate language and vision tasks. Despite their significance in advancing the field of Artificial Intelligence Generated Content (AIGC), their potential in intelligent audio content creation remains unexplored. In this work, we tackle the problem of creating audio content with storylines encompassing speech, music, and sound effects, guided by text instructions. We present WavJourney, a system that leverages LLMs to connect various audio models for audio content generation. Given a text description of an auditory scene, WavJourney first prompts LLMs to generate a structured script dedicated to audio storytelling. The audio script incorporates diverse audio elements, organized based on their spatio-temporal relationships. As a conceptual representation of audio, the audio script provides an interactive and interpretable rationale for human engagement. Afterward, the audio script is fed into a script compiler, converting it into a computer program. Each line of the program calls a task-specific audio generation model or computational operation function (e.g., concatenate, mix). The computer program is then executed to obtain an explainable solution for audio generation. We demonstrate the practicality of WavJourney across diverse real-world scenarios, including science fiction, education, and radio play. The explainable and interactive design of WavJourney fosters human-machine co-creation in multi-round dialogues, enhancing creative control and adaptability in audio production. WavJourney audiolizes the human imagination, opening up new avenues for creativity in multimedia content creation.

Many existing audio processing and generation models rely on task-specific architectures, resulting in fragmented development efforts and limited extensibility. It is therefore promising to design a unified framework capable of handling multiple tasks, while providing robust instruction and audio understanding and high-quality audio generation. This requires a compatible paradigm design, a powerful backbone, and a high-fidelity audio reconstruction module. To meet these requirements, this technical report introduces QuarkAudio, a decoder-only autoregressive (AR) LM-based generative framework that unifies multiple tasks. The framework includes a unified discrete audio tokenizer, H-Codec, which incorporates self-supervised learning (SSL) representations into the tokenization and reconstruction process. We further propose several improvements to H-Codec, such as a dynamic frame-rate mechanism and extending the audio sampling rate to 48 kHz. QuarkAudio unifies tasks by using task-specific conditional information as the conditioning sequence of the decoder-only LM, and predicting discrete target audio tokens in an AR manner. The framework supports a wide range of audio processing and generation tasks, including speech restoration (SR), target speaker extraction (TSE), speech separation (SS), voice conversion (VC), and language-queried audio source separation (LASS). In addition, we extend downstream tasks to universal free-form audio editing guided by natural language instructions (including speech semantic editing and audio event editing). Experimental results show that H-Codec achieves high-quality audio reconstruction with a low frame rate, improving both the efficiency and performance of downstream audio generation, and that QuarkAudio delivers competitive or comparable performance to state-of-the-art task-specific or multi-task systems across multiple tasks.

Real-time text-driven joint audio-video avatar generation requires jointly synthesizing portrait video and speech with high fidelity and precise synchronization, yet existing audio-visual diffusion models remain too slow for interactive use and often degrade noticeably after aggressive acceleration. We present Hallo-Live, a streaming framework for joint audio-visual avatar generation that combines asynchronous dual-stream diffusion with human-centric preference-guided distillation. To reduce articulation lag in causal generation, we introduce Future-Expanding Attention, which allows each video block to access synchronous audio together with a short horizon of future phonetic cues. To mitigate the quality loss of few-step distillation, we further propose Human-Centric Preference-Guided DMD (HP-DMD), which reweights training samples using rewards from visual fidelity, speech naturalness, and audio-visual synchronization. On two NVIDIA H200 GPUs, Hallo-Live runs at 20.38 FPS with 0.94 seconds latency, yielding 16.0x higher throughput and 99.3x lower latency than the teacher model Ovi. Despite this speedup, it retains strong generation quality, reaching comparable VideoAlign overall score and Sync Confidence score while outperforming other accelerated baselines in the overall quality-efficiency trade-off. Qualitative results further show robust generalization across photorealistic, multi-speaker, and stylized scenarios. To the best of our knowledge, Hallo-Live is the first framework to combine streaming dual-stream diffusion with preference-guided distillation for real-time, text-driven audio-visual generation.

Precise emotion control in audio-driven talking heads remains a challenge due to the reliance on implicit emotion regulation in existing systems, which often leads to indirect and insufficient control. Additionally, training with explicit emotion-related losses across the entire motion space poses significant difficulties due to the inherent trade-off between accurate lip synchronization and fine-grained emotion control. In this paper, we reveal a key finding: although emotional cues are distributed throughout the motion space, concentrating discriminative supervision on less-principal components achieves a better emotion-lip synchronization balance, as principal components mainly encode high-energy articulation and pose variations. Building on this insight, we propose Xemo-Talker, which first learns a neutral speech-to-motion mapping for stable articulation and lip synchronization, and then introduces a lightweight emotion branch guided by less-principal subspace supervision. To enhance emotion control, we design a Tri-Loss consisting of inter-class separation, intra-class compactness, and less-principal contrastive learning. Given an audio input, a reference image, and an emotion label, Xemo-Talker achieves state-of-the-art emotion classification accuracy while maintaining competitive lip synchronization and high inference efficiency, with performance approaching that measured on real videos.The source code is publicly available at https://github.com/chaolongy/Xemo-Talker.

Instruct Text-to-Speech (InstructTTS) leverages natural language descriptions as style prompts to guide speech synthesis. However, existing InstructTTS methods mainly rely on a direct combination of audio-related labels or their diverse rephrasings, making it difficult to handle flexible, high-level instructions. Such rigid control is insufficient for users such as content creators who wish to steer generation with descriptive instructions. To address these constraints, we introduce OV-InstructTTS, a new paradigm for open-vocabulary InstructTTS. We propose a comprehensive solution comprising a newly curated dataset, OV-Speech, and a novel reasoning-driven framework. The OV-Speech dataset pairs speech with open-vocabulary instructions, each augmented with a reasoning process that connects high-level instructions to acoustic features. The reasoning-driven framework infers emotional, acoustic, and paralinguistic information from open-vocabulary instructions before synthesizing speech. Evaluations show that this reasoning-driven approach significantly improves instruction-following fidelity and speech expressiveness. We believe this work can inspire the next user-friendly InstructTTS systems with stronger generalization and real-world applicability. The dataset and demos are publicly available on our project page.

Using a text description as prompt to guide the generation of text or images (e.g., GPT-3 or DALLE-2) has drawn wide attention recently. Beyond text and image generation, in this work, we explore the possibility of utilizing text descriptions to guide speech synthesis. Thus, we develop a text-to-speech (TTS) system (dubbed as PromptTTS) that takes a prompt with both style and content descriptions as input to synthesize the corresponding speech. Specifically, PromptTTS consists of a style encoder and a content encoder to extract the corresponding representations from the prompt, and a speech decoder to synthesize speech according to the extracted style and content representations. Compared with previous works in controllable TTS that require users to have acoustic knowledge to understand style factors such as prosody and pitch, PromptTTS is more user-friendly since text descriptions are a more natural way to express speech style (e.g., ''A lady whispers to her friend slowly''). Given that there is no TTS dataset with prompts, to benchmark the task of PromptTTS, we construct and release a dataset containing prompts with style and content information and the corresponding speech. Experiments show that PromptTTS can generate speech with precise style control and high speech quality. Audio samples and our dataset are publicly available.

Controlling text-to-speech (TTS) systems to synthesize speech with the prosodic characteristics expected by users has attracted much attention. To achieve controllability, current studies focus on two main directions: (1) using reference speech as prosody prompt to guide speech synthesis, and (2) using natural language descriptions to control the generation process. However, finding reference speech that exactly contains the prosody that users want to synthesize takes a lot of effort. Description-based guidance in TTS systems can only determine the overall prosody, which has difficulty in achieving fine-grained prosody control over the synthesized speech. In this paper, we propose DrawSpeech, a sketch-conditioned diffusion model capable of generating speech based on any prosody sketches drawn by users. Specifically, the prosody sketches are fed to DrawSpeech to provide a rough indication of the expected prosody trends. DrawSpeech then recovers the detailed pitch and energy contours based on the coarse sketches and synthesizes the desired speech. Experimental results show that DrawSpeech can generate speech with a wide variety of prosody and can precisely control the fine-grained prosody in a user-friendly manner. Our implementation and audio samples are publicly available.

Chain-of-thought (CoT) prompting has been extended to large audio-language models (LALMs) to elicit reasoning, yet enhancing its effectiveness without training remains challenging. We study inference-time model steering as a training-free approach to improve LALM reasoning. We introduce three strategies using diverse information sources and evaluate them across four LALMs and four benchmarks. Results show general accuracy gains up to 4.4% over CoT prompting. Notably, we identify a cross-modal transfer where steering vectors derived from few text samples effectively guide speech-based reasoning, demonstrating high data efficiency. We also examine hyperparameter sensitivity to understand the robustness of these approaches. Our findings position model steering as a practical direction for strengthening LALM reasoning.

Masked modeling framework has shown promise in co-speech motion generation. However, it struggles to identify semantically significant frames for effective motion masking. In this work, we propose a speech-queried attention-based mask modeling framework for co-speech motion generation. Our key insight is to leverage motion-aligned speech features to guide the masked motion modeling process, selectively masking rhythm-related and semantically expressive motion frames. Specifically, we first propose a motion-audio alignment module (MAM) to construct a latent motion-audio joint space. In this space, both low-level and high-level speech features are projected, enabling motion-aligned speech representation using learnable speech queries. Then, a speech-queried attention mechanism (SQA) is introduced to compute frame-level attention scores through interactions between motion keys and speech queries, guiding selective masking toward motion frames with high attention scores. Finally, the motion-aligned speech features are also injected into the generation network to facilitate co-speech motion generation. Qualitative and quantitative evaluations confirm that our method outperforms existing state-of-the-art approaches, successfully producing high-quality co-speech motion.

Audio-based video object segmentation aims to locate and segment objects in videos conditioned on audio cues, requiring precise understanding of both appearance and motion. Recent audio-driven video segmentation methods extend MLLMs by fusing audio and visual features for end-to-end localization. Despite their promise, these approaches are computationally intensive, struggle with aligning temporal audio cues to dynamic video content, and depend on large paired audio-video datasets. To address these challenges, we present ASR-SaSaSa2VA, a resource-efficient framework for audio-guided video segmentation. The key idea is to convert audio inputs into textual motion descriptions via automatic speech recognition (ASR) models and then leverage pre-trained text-based referring video segmentation models (e.g., SaSaSa2VA) for pixel-level predictions. To further enhance robustness, we incorporate a no-target expression detection module, implemented by a fine-tuned audio-based MLLM, which filters out audio clips that do not refer to any target object. This design allows the system to exploit strong pre-trained models while effectively handling ambiguous or irrelevant audio inputs. Our approach achieves a final score of 80.7 in the 5th PVUW Challenge (MeViS-v2-Audio track), earning the second-place ranking.

Despite advances in text and visual generation, creating coherent long-form audio narratives remains challenging. Existing frameworks often exhibit limitations such as mismatched character settings with voice performance, insufficient self-correction mechanisms, and limited human interactivity. To address these challenges, we propose AuDirector, a self-reflective closed-loop multi-agent framework. Specifically, it involves an Identity-Aware Pre-production mechanism that transforms narrative texts into character profiles and utterance-level emotional instructions to retrieve suitable voice candidates and guide expressive speech synthesis, thereby promoting context-aligned voice adaptation. To enhance quality, a Collaborative Synthesis and Correction module introduces a closed-loop self-correction mechanism to systematically audit and regenerate defective audio components. Furthermore, a Human-Guided Interactive Refinement module facilitates user control by interpreting natural language feedback to interactively refine the underlying scripts. Experiments demonstrate that AuDirector achieves superior performance compared to state-of-the-art baselines in structural coherence, emotional expressiveness, and acoustic fidelity. Audio samples can be found at https://anonymous-itsh.github.io/.

The study of the attention mechanism has sparked interest in many fields, such as language modeling and machine translation. Although its patterns have been exploited to perform different tasks, from neural network understanding to textual alignment, no previous work has analysed the encoder-decoder attention behavior in speech translation (ST) nor used it to improve ST on a specific task. In this paper, we fill this gap by proposing an attention-based policy (EDAtt) for simultaneous ST (SimulST) that is motivated by an analysis of the existing attention relations between audio input and textual output. Its goal is to leverage the encoder-decoder attention scores to guide inference in real time. Results on en->{de, es} show that the EDAtt policy achieves overall better results compared to the SimulST state of the art, especially in terms of computational-aware latency.

Due to their significance in human communication, the automatic generation of co-speech gestures in artificial embodied agents has received a lot of attention. Although modern deep learning approaches can generate realistic-looking conversational gestures from spoken language, they often lack the ability to convey meaningful information and generate contextually appropriate gestures. This paper presents an augmented approach to the generation of co-speech gestures that additionally takes into account given form and meaning features for the gestures. Our framework effectively acquires this information from a small corpus with rich semantic annotations and a larger corpus without such information. We provide an analysis of the effects of distinctive feature targets and we report on a human rater evaluation study demonstrating that our framework achieves semantic coherence and person perception on the same level as human ground truth behavior. We make our data pipeline and the generation framework publicly available.

Attention is the core mechanism of today's most used architectures for natural language processing and has been analyzed from many perspectives, including its effectiveness for machine translation-related tasks. Among these studies, attention resulted to be a useful source of information to get insights about word alignment also when the input text is substituted with audio segments, as in the case of the speech translation (ST) task. In this paper, we propose AlignAtt, a novel policy for simultaneous ST (SimulST) that exploits the attention information to generate source-target alignments that guide the model during inference. Through experiments on the 8 language pairs of MuST-C v1.0, we show that AlignAtt outperforms previous state-of-the-art SimulST policies applied to offline-trained models with gains in terms of BLEU of 2 points and latency reductions ranging from 0.5s to 0.8s across the 8 languages.

Text anonymization is essential for responsibly developing and deploying AI in high-stakes domains such as healthcare, social services, and law. In this work, we propose a novel methodology for privacy-preserving synthetic text generation that leverages the principles of de-identification and the Hiding In Plain Sight (HIPS) theory. Our approach introduces entity-aware control codes to guide controllable generation using either in-context learning (ICL) or prefix tuning. The ICL variant ensures privacy levels consistent with the underlying de-identification system, while the prefix tuning variant incorporates a custom masking strategy and loss function to support scalable, high-quality generation. Experiments on legal and clinical datasets demonstrate that our method achieves a strong balance between privacy protection and utility, offering a practical and effective solution for synthetic text generation in sensitive domains.

The success of large language models (LLMs) has prompted efforts to integrate speech and audio data, aiming to create general foundation models capable of processing both textual and non-textual inputs. Recent advances, such as GPT-4o, highlight the potential for end-to-end speech LLMs, which preserves non-semantic information and world knowledge for deeper speech understanding. To guide the development of speech LLMs, we propose a five-level roadmap, ranging from basic automatic speech recognition (ASR) to advanced superhuman models capable of integrating non-semantic information with abstract acoustic knowledge for complex tasks. Moreover, we design a benchmark, SAGI Bechmark, that standardizes critical aspects across various tasks in these five levels, uncovering challenges in using abstract acoustic knowledge and completeness of capability. Our findings reveal gaps in handling paralinguistic cues and abstract acoustic knowledge, and we offer future directions. This paper outlines a roadmap for advancing speech LLMs, introduces a benchmark for evaluation, and provides key insights into their current limitations and potential.