124 lines
5.1 KiB
Markdown
124 lines
5.1 KiB
Markdown
# ComfyUI Audio Nodes
|
|
|
|
ComfyUI-EdgeTTS is a powerful text-to-speech node for ComfyUI, leveraging Microsoft's Edge TTS capabilities. It enables seamless conversion of text into natural-sounding speech, supporting multiple languages and voices. Ideal for enhancing user interactions, this node is easy to integrate and customize, making it perfect for various applications.
|
|
|
|
https://github.com/user-attachments/assets/a5b9165b-a413-49fd-989e-0ef3141afce7
|
|
|
|
## Updates
|
|
- V1.2.2 (2026-01-25) - Voice ID update requirments [update log](update.md#v121-2026125).
|
|
- V1.2.1 (2025-07-23) - Voice ID update & Bug Fixed [update log](update.md#v121-2025723).
|
|
- V1.2.0 (2025-06-20) - Simplified voice display format, improved performance with lazy loading and caching, and reduced memory usage. For more information, please see the [update log](update.md#v120-2025620).
|
|
|
|
- V1.1.0 (2025-01-24) - Added 19 new languages and 38 new voices, with more detailed characteristics for existing Chinese voices. For more information, please see the [update log](update.md#v110-2025124).
|
|
|
|
## Features
|
|
|
|
### Edge TTS Node
|
|
- **Edge TTS**: Convert text to speech using Microsoft Edge TTS
|
|
- Multiple languages and voices support
|
|
- Adjustable speech rate and pitch
|
|
- High-quality voice synthesis
|
|
- Configurable via config.json
|
|
|
|
### Speech to Text Node
|
|
- **Whisper STT**: High-accuracy speech recognition
|
|
- Multiple language support with auto-detection
|
|
- Multiple model sizes (tiny to large)
|
|
- Supports ComfyUI audio format
|
|
- Language detection confidence reporting
|
|
|
|
### Audio File Node
|
|
- **Save Audio**: Export audio files
|
|
- Supports WAV, MP3, FLAC formats
|
|
- Quality presets (high/medium/low)
|
|
- Custom file naming and paths
|
|
- Automatic file numbering
|
|
|
|
## Installation
|
|
|
|
### Method 1. install on ComfyUI-Manager, search `Comfyui-EdgeTTS` and install
|
|
install requirment.txt in the ComfyUI-EdgeTTS folder
|
|
```bash
|
|
./ComfyUI/python_embeded/python -m pip install -r requirements.txt
|
|
```
|
|
|
|
### Method 2. Clone this repository to your ComfyUI custom_nodes folder:
|
|
```bash
|
|
cd ComfyUI/custom_nodes
|
|
git clone https://github.com/1038lab/ComfyUI-EdgeTTS.git
|
|
```
|
|
install requirment.txt in the ComfyUI-EdgeTTS folder
|
|
```bash
|
|
./ComfyUI/python_embeded/python -m pip install -r requirements.txt
|
|
```
|
|
## Requirements
|
|
- Python packages (see `requirements.txt`)
|
|
- FFmpeg on system `PATH` (required by Whisper STT)
|
|
- Example (Windows PowerShell): `$env:Path += ";F:\\FFmpeg\\bin"` then restart ComfyUI
|
|
- CUDA compatible GPU (optional, for faster Whisper processing)
|
|
|
|
> [!NOTE]
|
|
> ### Important: torchaudio 2.9+ requires torchcodec
|
|
> If you use `torch/torchaudio` 2.9 or newer, `torchaudio.load/save` requires `torchcodec`.
|
|
|
|
Recommended fix (match PyTorch 2.9.x):
|
|
```bash
|
|
./ComfyUI/python_embeded/python -m pip uninstall -y torchcodec
|
|
./ComfyUI/python_embeded/python -m pip install --no-cache-dir "torchcodec==0.9"
|
|
```
|
|
|
|
### Text to Speech
|
|
1. Add Edge TTS node to workflow
|
|
2. Input text and select voice
|
|
3. Adjust speed and pitch if needed
|
|
4. Connect to Save Audio node for export
|
|
|
|
|
|
|
|
### Speech to Text
|
|
1. Add Whisper STT node
|
|
2. Connect audio input
|
|
3. Select model size and language (or auto-detect)
|
|
4. Run to get transcription
|
|
|
|
|
|
|
|
## Supported Voices
|
|
|
|
| Language | Female Voices | Male Voices |
|
|
|----------|--------------|-------------|
|
|
| Chinese-Mainland | XiaoXiao (News, Novel, Warm), XiaoYi (Cartoon, Novel, Lively) | Yunjian (Sports, Novel, Passion), Yunxi (Novel, Lively), Yunxia (Cartoon, Novel), Yunyang (News, Professional) |
|
|
| Chinese-Cantonese | HiuGaai (Friendly), HiuMaan (Friendly) | WanLung (Friendly) |
|
|
| Chinese-Taiwan | HsiaoChen (Friendly), HsiaoYu (Friendly) | YunJhe (Friendly) |
|
|
| English-US | Jenny (Friendly), Aria (Positive), Ana (Cute), Michelle (Friendly) | Guy (Passion), Christopher (Authority), Eric (Rational), Roger (Lively), Steffan (Rational) |
|
|
| English-GB | Libby (Friendly), Maisie (Friendly), Sonia (Friendly) | Ryan (Friendly), Thomas (Friendly) |
|
|
| English-AU | Natasha (Friendly) | William (Friendly) |
|
|
| Japanese | Nanami (Friendly) | Keita (Friendly) |
|
|
| Korean | SunHi (Friendly) | InJoon (Friendly), Hyunsu (Multilingual) |
|
|
| French-FR | Denise (Friendly), Eloise (Friendly), Vivienne (Multilingual) | Henri (Friendly), Remy (Multilingual) |
|
|
| French-CA | Sylvie (Friendly) | Jean (Friendly), Antoine (Friendly) |
|
|
| German-DE | Katja (Friendly), Amala (Friendly), Seraphina (Multilingual) | Conrad (Friendly), Killian (Friendly), Florian (Multilingual) |
|
|
|
|
More voices available in config.json, including voices for:
|
|
- German (AT/CH)
|
|
- Spanish (ES/MX)
|
|
- Russian
|
|
- Italian
|
|
- Portuguese (BR/PT)
|
|
- Dutch
|
|
- Polish
|
|
- Turkish
|
|
- Arabic
|
|
- Hindi
|
|
- Indonesian
|
|
- Vietnamese
|
|
- Thai
|
|
- Ukrainian
|
|
And many more...
|
|
|
|
Each language provides at least one male and female voice option, allowing you to choose different voice styles based on your needs.
|
|
|
|
## Credits
|
|
- Edge TTS: [Microsoft Edge TTS](https://github.com/rany2/edge-tts)
|
|
- Whisper: [OpenAI Whisper](https://github.com/openai/whisper)
|