|
Download README.md from oddadmix/tiny-voca-arabic: direct link, hf CLI and curl.
- Browser
- Download file 7.81 kB
-
https://huggingface.co/oddadmix/tiny-voca-arabic/resolve/main/README.md
- Command line
-
hf download hf://oddadmix/tiny-voca-arabic/README.md
-
curl -L -o README.md https://huggingface.co/oddadmix/tiny-voca-arabic/resolve/main/README.md
7.81 kB
| language: [ar] | |
| license: apache-2.0 | |
| library_name: onnxruntime | |
| pipeline_tag: text-to-speech | |
| tags: | |
| - text-to-speech | |
| - tts | |
| - speech-synthesis | |
| - arabic | |
| - msa | |
| - modern-standard-arabic | |
| - tiny-voca | |
| - onnx | |
| - cpu | |
| - small-model | |
| - trained-from-scratch | |
| datasets: [oddadmix/msa-duplex-assistant-tts] | |
| # tiny-voca-arabic | |
| A **4.5M-parameter** Modern Standard Arabic (فصحى) text-to-speech model, trained **from scratch** with the | |
| [tiny-voca](https://huggingface.co/n4ze3m/tiny-voca) architecture and a new Arabic phonetic front end. | |
| The waveform generator is included: text in, 24 kHz audio out, about 200× real time on two CPU threads. | |
| > **Based on [tiny-voca](https://huggingface.co/n4ze3m/tiny-voca) by Muhammed Nazeem ([@n4ze3m](https://huggingface.co/n4ze3m)).** | |
| > This model uses the original tiny-voca architecture and training code | |
| > ([github.com/n4ze3m/tiny-voca](https://github.com/n4ze3m/tiny-voca), Apache-2.0). No tiny-voca weights were reused: | |
| > it was trained from scratch on Arabic data. The changes are listed under [Changes from tiny-voca](#changes-from-tiny-voca). | |
| [Demo](https://huggingface.co/spaces/oddadmix/tiny-voca-arabic) · | |
| [Dataset](https://huggingface.co/datasets/oddadmix/msa-duplex-assistant-tts) | |
| | | | | |
| |---|---| | |
| | Parameters | 4,531,723 (18 MB PyTorch, 21 MB ONNX) | | |
| | Voice | one synthetic MSA voice (`a000`) | | |
| | Audio | 24 kHz mono | | |
| | Training data | 45,737 utterances, 89 h | | |
| | Training | 104k steps, 2.6 h on one GPU | | |
| ## Samples | |
| | Text | Audio | | |
| |---|---| | |
| | مرحبًا بكم، هذا نموذج صغير لتحويل النص إلى كلام باللغة العربية الفصحى. | [sample-0.wav](samples/sample-0.wav) | | |
| | تقع مدينة القاهرة على ضفاف نهر النيل، وهي من أكبر المدن في العالم العربي. | [sample-1.wav](samples/sample-1.wav) | | |
| | هل يمكنك أن تخبرني عن حالة الطقس غدًا؟ أريد أن أخرج في نزهة مع أصدقائي. | [sample-2.wav](samples/sample-2.wav) | | |
| | يبلغ عدد سكان المدينة 2,500,000 نسمة، ونسبة الشباب فيها 35%. | [sample-3.wav](samples/sample-3.wav) | | |
| | العلم نور، والجهل ظلام. ومن جدّ وجد، ومن زرع حصد. | [sample-4.wav](samples/sample-4.wav) | | |
| ## How it works | |
| 1. **Numbers** are read as MSA words (`code/numbers.py`). | |
| 2. **Diacritization** with [CATT](https://github.com/abjadai/catt) (encoder-decoder, Apache-2.0). | |
| 3. **Phonemes** from a rule-based MSA converter (`code/arabic_frontend.py`), 49 symbols: | |
| sun-letter assimilation, hamzat al-wasl elision and helping vowels, long vowels shortened before | |
| wasl, shadda as a doubled consonant, tanween, tāʾ marbūṭa (`t` inside a phrase, dropped at a pause), | |
| pausal forms at punctuation, emphatic backing of /a/ next to ص ض ط ظ ق (and after خ غ ر), dark lām in | |
| الله, and stress from syllable weight. | |
| 4. **Acoustic model + synthesizer**: the tiny-voca network predicts durations, pauses, pitch, voicing and | |
| energy, then a source-filter synthesizer makes the waveform. There is no separate neural vocoder. | |
| **Case endings.** Speakers often drop mid-sentence iʿrāb. While aligning the training data, each | |
| mid-phrase case ending was kept or removed per clip according to which version the audio supports | |
| (a CTC likelihood comparison), so the model learned from what was actually said. About half were dropped. | |
| ## Evaluation | |
| Whisper large-v3 transcripts of generated speech against the input text (diacritics and punctuation removed). | |
| Held-out sets come from dialogues not seen in training; the new sentences appear in no split. | |
| | Set | Clips | CER | WER | CER of the original recordings | | |
| |---|---|---|---|---| | |
| | Validation | 25 | 0.5% | 1.2% | 1.0% | | |
| | Test | 25 | 1.8% | 2.4% | 2.1% | | |
| | New sentences | 8 | 0.3% | 1.4% | — | | |
| Most remaining errors are Whisper writing spoken numbers as digits. Generated durations are within 3% of | |
| the recordings and pitch variation matches (3.6 semitones). | |
| ## Usage (ONNX Runtime) | |
| ```bash | |
| pip install onnxruntime numpy soundfile tqdm | |
| hf download oddadmix/tiny-voca-arabic --local-dir tiny-voca-arabic | |
| ``` | |
| The front end in `code/` uses the CATT diacritizer; the | |
| [demo Space](https://huggingface.co/spaces/oddadmix/tiny-voca-arabic/tree/main) contains it ready to run | |
| (`app.py`, `tiny_voca_ar/`, `catt_tashkeel/`). Minimal call: | |
| ```python | |
| import numpy as np, onnxruntime as ort, soundfile as sf | |
| from tiny_voca_ar.numbers import expand_numbers | |
| from tiny_voca_ar.text import phonemize | |
| session = ort.InferenceSession("tiny-voca-arabic/onnx/model.onnx", providers=["CPUExecutionProvider"]) | |
| phones = phonemize(expand_numbers("مرحبًا بكم في عالم الذكاء الاصطناعي.")) | |
| one = lambda v: np.array([v], dtype=np.float32) | |
| audio, durations = session.run(None, { | |
| "tokens": np.array([phones.ids], dtype=np.int64), "stress": np.array([phones.stress], dtype=np.int64), | |
| "voice": np.array([0], dtype=np.int64), "speed": one(1.0), "pitch_shift": one(0.0), "pitch_range": one(1.0)}) | |
| sf.write("hello.wav", audio[0], 24000) | |
| ``` | |
| Inputs: `tokens`/`stress` int64 `[1, N]`, `voice` int64 `[1]`, `speed` (1 = as predicted, larger is | |
| faster), `pitch_shift` (added to log F0; 0.058 ≈ one semitone), `pitch_range` (scales intonation). Read | |
| long text one sentence at a time (at most 500 phonemes / 32 s per call) and join the audio. | |
| The PyTorch checkpoint `tiny-voca-arabic.pt` loads with the tiny-voca code | |
| (`tiny_voca.infer.load_model`) using the Arabic `text.py` / `arabic_frontend.py` from `code/`. | |
| ## Limitations | |
| - One voice, Modern Standard Arabic only; dialectal text is read with MSA rules. | |
| - Pronunciation follows the automatic diacritization. When CATT picks the wrong reading (for example | |
| نَمَوذج instead of نُموذج), the model says it. Add tashkeel yourself to fix a word. | |
| - Latin script, emoji and symbols other than numbers and % are skipped. | |
| - The ص/ط/ض/ظ emphatics are well covered, but the dark lām only appears in الله. | |
| ## Training data | |
| [oddadmix/msa-duplex-assistant-tts](https://huggingface.co/datasets/oddadmix/msa-duplex-assistant-tts) | |
| (batches 0–8): single-speaker synthetic MSA speech with scripted assistant dialogue turns. Of 49,793 clips, | |
| 45,737 passed alignment quality checks. The voice is synthetic; please tell listeners when audio | |
| they hear was generated. | |
| ## License | |
| Apache-2.0. | |
| ## Changes from tiny-voca | |
| The architecture is unchanged from tiny-voca 903f940: same network, synthesizer and losses, at the same size. The changes are: | |
| - An Arabic front end that replaces the English phonemizer: CATT diacritization, MSA phoneme rules, a | |
| 49-symbol inventory and a number reader. | |
| - One voice instead of two (F2/M2). | |
| - Aligner label mapping for the Arabic emphatics, and per-clip selection of spoken case endings during annotation. | |
| - Training on 89 h of Arabic speech with packed memory-mapped features, plus an Arabic evaluation that uses | |
| Whisper large-v3 CER. | |
| ## Credits | |
| - **Original model, architecture and training code:** [tiny-voca](https://huggingface.co/n4ze3m/tiny-voca) | |
| by Muhammed Nazeem ([@n4ze3m](https://huggingface.co/n4ze3m)), Apache-2.0 | |
| ([code](https://github.com/n4ze3m/tiny-voca)). | |
| - **Diacritizer:** [CATT](https://github.com/abjadai/catt) by Abjad AI, Apache-2.0. | |
| ## Citation | |
| If you use this model, please cite: | |
| ```bibtex | |
| @software{oddadmix2026tinyvocaarabic, | |
| author = {oddadmix}, | |
| title = {tiny-voca-arabic: a 4.5M-parameter Modern Standard Arabic text-to-speech model}, | |
| year = {2026}, | |
| url = {https://huggingface.co/oddadmix/tiny-voca-arabic}, | |
| license = {Apache-2.0} | |
| } | |
| ``` | |
| Architecture and training code are from [tiny-voca](https://huggingface.co/n4ze3m/tiny-voca) by Muhammed Nazeem (see Credits). | |