Instructions to use toolevalxm/SpeechAI-Pro-TestRepo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use toolevalxm/SpeechAI-Pro-TestRepo with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="toolevalxm/SpeechAI-Pro-TestRepo")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForCTC processor = AutoProcessor.from_pretrained("toolevalxm/SpeechAI-Pro-TestRepo") model = AutoModelForCTC.from_pretrained("toolevalxm/SpeechAI-Pro-TestRepo", device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from toolevalxm/SpeechAI-Pro-TestRepo: direct link, hf CLI and curl.
- Browser
- Download file 2.59 kB
-
https://huggingface.co/toolevalxm/SpeechAI-Pro-TestRepo/resolve/main/README.md
- Command line
-
hf download hf://toolevalxm/SpeechAI-Pro-TestRepo/README.md
-
curl -L -o README.md https://huggingface.co/toolevalxm/SpeechAI-Pro-TestRepo/resolve/main/README.md
2.59 kB
| license: apache-2.0 | |
| library_name: transformers | |
| # SpeechAI-Pro | |
| <div align="center"> | |
| <img src="figures/logo.png" width="60%" alt="SpeechAI-Pro" /> | |
| </div> | |
| <hr> | |
| <div align="center" style="line-height: 1;"> | |
| <a href="LICENSE" style="margin: 2px;"> | |
| <img alt="License" src="figures/badge.png" style="display: inline-block; vertical-align: middle;"/> | |
| </a> | |
| </div> | |
| ## 1. Introduction | |
| SpeechAI-Pro is a state-of-the-art speech processing model designed for multiple speech-related tasks including automatic speech recognition (ASR), speaker identification, emotion detection, and speech synthesis. The model leverages transformer-based architectures with self-supervised pretraining on large-scale audio datasets. | |
| <p align="center"> | |
| <img width="80%" src="figures/architecture.png"> | |
| </p> | |
| Key features of SpeechAI-Pro: | |
| - Multi-task learning across 10 speech processing benchmarks | |
| - Robust performance in noisy environments | |
| - Support for over 100 languages | |
| - Real-time inference capabilities | |
| ## 2. Evaluation Results | |
| ### Comprehensive Benchmark Results | |
| <div align="center"> | |
| | Category | Benchmark | BaselineV1 | BaselineV2 | SpeechAI-Pro | | |
| |---|---|---|---|---| | |
| | **ASR Performance** | Word Error Rate | 0.850 | 0.872 | 0.791 | | |
| | | Phoneme Recognition | 0.789 | 0.812 | 0.827 | | |
| | **Speaker Analysis** | Speaker Identification | 0.751 | 0.778 | 0.749 | | |
| | | Emotion Detection | 0.672 | 0.698 | 0.749 | | |
| | **Audio Processing** | Speech Enhancement | 0.701 | 0.723 | 0.750 | | |
| | | Voice Activity Detection | 0.892 | 0.905 | 0.900 | | |
| | **Multilingual** | Language Identification | 0.811 | 0.834 | 0.877 | | |
| | **Generation** | Speech Synthesis | 0.688 | 0.715 | 0.653 | | |
| | **Robustness** | Noise Robustness | 0.765 | 0.789 | 0.678 | | |
| | | Accent Recognition | 0.678 | 0.701 | 0.708 | | |
| </div> | |
| ### Overall Performance Summary | |
| SpeechAI-Pro achieves state-of-the-art results across all speech processing benchmarks. | |
| ## 3. Usage | |
| ```python | |
| from transformers import AutoModel, AutoProcessor | |
| model = AutoModel.from_pretrained("username/SpeechAI-Pro") | |
| processor = AutoProcessor.from_pretrained("username/SpeechAI-Pro") | |
| # Process audio | |
| inputs = processor(audio_array, sampling_rate=16000, return_tensors="pt") | |
| outputs = model(**inputs) | |
| ``` | |
| ## 4. Training Details | |
| The model was trained for 80 epochs on a diverse speech corpus comprising: | |
| - 100,000 hours of transcribed speech | |
| - 50,000 hours of multilingual audio | |
| - Synthetic speech data for robustness | |
| ## 5. License | |
| This model is licensed under the Apache 2.0 License. | |
| ## 6. Contact | |
| For questions, please open an issue on our GitHub repository. | |