| --- |
| title: BlueTTS - Multilingual TTS |
| emoji: 🔊 |
| colorFrom: blue |
| colorTo: indigo |
| sdk: gradio |
| sdk_version: 6.9.0 |
| app_file: app.py |
| pinned: false |
| python_version: "3.12" |
| --- |
| |
| # BlueTTS — Multilingual text-to-speech |
|
|
| **This Space:** [https://huggingface.co/spaces/notmax123/Blue](https://huggingface.co/spaces/notmax123/Blue) |
|
|
| Interactive demo for **[BlueTTS](https://github.com/maxmelichov/BlueTTS)**: fast on-device synthesis with **ONNX Runtime**. Type text, pick a language and voice, and generate audio in the browser. Project site: [lightbluetts.com](https://lightbluetts.com/). |
|
|
| This Space was duplicated from [thewh1teagle/lightblue-tts-hebrew](https://huggingface.co/spaces/thewh1teagle/lightblue-tts-hebrew) and adapted for the current BlueTTS ONNX bundle. |
|
|
| ## Languages |
|
|
| - **English, Spanish, German, Italian** — phonemization with **espeak-ng** (via phonemizer), aligned with the multilingual tokenizer in the app. |
| - **Hebrew** — grapheme-to-phoneme via [Renikud](https://huggingface.co/thewh1teagle/renikud) (ONNX). |
|
|
| ## Voices |
|
|
| Two built-in speaker styles (pre-extracted style JSON): |
|
|
| | Voice | File | |
| | ------ | ----------------------- | |
| | Female | `voices/female1.json` | |
| | Male | `voices/male1.json` | |
|
|
| ## Implementation note |
|
|
| This Space runs **inference from `app.py`** (vocabulary, phonemization, ONNX sessions). It does **not** install the upstream [`blue`](https://github.com/maxmelichov/BlueTTS) Python package from GitHub (that package name replaced the old `lightblue-onnx` wheel and would conflict with pip’s expected name in a flat `requirements.txt`). |
|
|
| ## Model weights |
|
|
| ONNX checkpoints are **not** stored in this Space repo (file-size limits). On first startup, `download_models.py` fetches them from the public model repo **[notmax123/blue-onnx](https://huggingface.co/notmax123/blue-onnx)** into `./onnx_models/`. |
|
|
| **Training / finetuning / voice-export weights** (PyTorch / Safetensors): **[notmax123/blue](https://huggingface.co/notmax123/blue)**. |
|
|
| If you fork this Space and point downloads at a **private** repo, set a Hugging Face **write** token as the Space secret `HF_TOKEN` so authenticated downloads still work. |
|
|
| ## Links |
|
|
| - **This demo (Space):** [huggingface.co/spaces/notmax123/Blue](https://huggingface.co/spaces/notmax123/Blue) |
| - **ONNX weights:** [notmax123/blue-onnx](https://huggingface.co/notmax123/blue-onnx) |
| - **PyTorch / training weights:** [notmax123/blue](https://huggingface.co/notmax123/blue) |
| - **Source code:** [maxmelichov/BlueTTS on GitHub](https://github.com/maxmelichov/BlueTTS) |
| - **Hebrew G2P:** [thewh1teagle/renikud](https://huggingface.co/thewh1teagle/renikud) |
|
|
| ## Local development |
|
|
| Dependencies are listed in `requirements.txt`. With Python 3.12: |
|
|
| ```bash |
| pip install -r requirements.txt |
| python download_models.py # once, to populate onnx_models/ |
| python app.py # or: gradio will launch from app_file when run by HF |
| ``` |
|
|
| To refresh `requirements.txt` from **uv**: |
|
|
| ```bash |
| uv export --no-hashes --no-emit-project > requirements.txt |
| ``` |
|
|
| ## License |
|
|
| See the [BlueTTS](https://github.com/maxmelichov/BlueTTS) repository and the [blue-onnx](https://huggingface.co/notmax123/blue-onnx) model card for licensing of code and weights. |
|
|
