Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
A local implementation of the Kokoro Text-to-Speech model, featuring dynamic module loading, automatic dependency management, and a web interface.
| Date | Stars |
|---|---|
| 2026-07-31 | 324 |
| 2026-08-02 | 327 |
| 2026-08-06 | 327 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Kokoro TTS Local A local implementation of the Kokoro Text-to-Speech model, featuring dynamic module loading, automatic dependency management, and a web interface. ## Features - Local text-to-speech synthesis using the Kokoro-82M model - Multiple voice support with easy voice selection (54 voices available across 9 languages) - Automatic model and voice downloading from Hugging Face - **Offline mode support** - Run completely offline after initial setup - Phoneme output support and visualization - Interactive CLI and web interface - Voice listing functionality - Cross-platform support (Windows, Linux, macOS) - Real-time generation progress display - Multiple output formats (WAV, MP3, AAC) - Enhanced security and code quality features - Centralized configuration management - Comprehensive dependency validation - Memory management and optimization - Thread-safe operations for multi-user scenarios ## Prerequisites - Python 3.10–3.12 (**Python 3.13+ is not supported** — core dependencies such as `misaki` and `numpy<2.0` do not provide packages for 3.13 yet) - FFmpeg (optional, for MP3/AAC conversion) - CUDA-compatible GPU (optional, for faster generation) - Git (for version control and package management) ## Installation 1. Clone the repository and create a Python virtual environment (use Python 3.10–3.12): ```bash # Windows python -m venv venv .\venv\Scripts\activate # If your default python is 3.13+, use a supported version instead, e.g.: # py -3.12 -m venv venv # Linux/macOS python3 -m venv venv source venv/bin/activate # If your default python3 is 3.13+, use e.g.: python3.12 -m venv venv ``` 2. Install the project and its dependencies: ```bash pip install -e . ``` 3. (Optional) For Japanese voices (`jf_*`/`jm_*`), download the UniDic dictionary data (~1 GB, one-time). Without it, Japanese G2P fails: ```bash python -m unidic download ``` This adds the following console commands to your virtual environment: - `kokoro-tts` — command-line interface (`python -m kokoro_tts_local.tts_demo`) - `kokoro-tts-web` — Gradio web interface (`python -m kokoro_tts_local.gradio_interface`) - `kokoro-tts-chinese` — Mandarin CLI (`python -m kokoro_tts_local.chinese_tts_demo`) - `kokoro-tts-setup` — Mandarin model and voice setup - `kokoro-tts-check` — dependency diagnostics **Alternative Installation (Simplified):** For a simpler setup, you can also install the official Kokoro package directly: ```bash pip install kokoro soundfile apt-get install espeak-ng # On Linux # or brew install espeak # On macOS ``` 4. (Optional) For GPU acceleration, install PyTorch with CUDA support: ```bash # For CUDA 11.8 pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 # For CUDA 12.1 pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121 # For CUDA 12.6 pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126 # For CUDA 12.8 (for RTX 50-series cards) pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128 ``` You can verify CUDA support is enabled with: ```python import torch print(torch.cuda.is_available()) # Should print True if CUDA is available ``` The system will automatically download required models and voice files on first run. ## Docker Quick Start This project can be run in a CPU-first Docker setup with runtime model and voice downloads. ### Build and Run with Docker **Linux/macOS (bash/zsh):** ```bash docker build -t kokoro-tts-local:cpu . docker run --rm -it \ -p 127.0.0.1:7860:7860 \ -e KOKORO_TTS_USERNAME=admin -e KOKORO_TTS_PASSWORD=change-me \ -v kokoro-data:/data \ kokoro-tts-local:cpu ``` **Windows (PowerShell):** ```powershell docker build -t kokoro-tts-local:cpu . docker run --rm -it ` -p 127.0.0.1:7860:7860 ` -e KOKORO_TTS_USERNAME=admin -e KOKORO_TTS_PASSWORD=change-me ` -v kokoro-data:/data ` kokoro-tts-local:cpu ``` Open `http://localhost:7860
Excerpt of 19,110 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:02c3683b49e9f488, desc:text-to-speech