mirror of
https://github.com/vegu-ai/talemate.git
synced 2025-12-16 11:47:48 +01:00
docs
This commit is contained in:
@@ -2,6 +2,15 @@
|
||||
|
||||
Local zero shot voice cloning from .wav files.
|
||||
|
||||
!!! warning "FFmpeg Required"
|
||||
Chatterbox requires FFmpeg for audio processing. If you encounter errors like `Could not load libtorchcodec` or `FFmpeg version 8: Could not load this library`, you need to install FFmpeg.
|
||||
|
||||
**Windows:** Run `install-ffmpeg.bat` from the Talemate root directory.
|
||||
|
||||
**Linux/macOS:** Install FFmpeg using your system package manager (versions 4-8 supported).
|
||||
|
||||
See the [TTS Troubleshooting Guide](troubleshooting.md#ffmpeg-not-found) for more details.
|
||||
|
||||

|
||||
|
||||
##### Device
|
||||
|
||||
@@ -13,14 +13,18 @@ In 0.32.0 Talemate's TTS (Text-to-Speech) agent has been completely refactored t
|
||||
## Supported APIs
|
||||
|
||||
### Local APIs
|
||||
- **Kokoro** - Fastest generation with predefined voice models and mixing
|
||||
- **F5-TTS** - Fast voice cloning with occasional mispronunciations
|
||||
- **Chatterbox** - High-quality voice cloning (slower generation)
|
||||
- **[Kokoro](kokoro.md)** - Fastest generation with predefined voice models and mixing
|
||||
- **[F5-TTS](f5tts.md)** - Fast voice cloning with occasional mispronunciations
|
||||
- **[Chatterbox](chatterbox.md)** - High-quality voice cloning (slower generation)
|
||||
|
||||
### Remote APIs
|
||||
- **ElevenLabs** - Professional voice synthesis with voice cloning
|
||||
- **Google Gemini-TTS** - Google's text-to-speech service
|
||||
- **OpenAI** - OpenAI's TTS-1 and TTS-1-HD models
|
||||
- **[ElevenLabs](elevenlabs.md)** - Professional voice synthesis with voice cloning
|
||||
- **[Google Gemini-TTS](google.md)** - Google's text-to-speech service
|
||||
- **[OpenAI](openai.md)** - OpenAI's TTS-1 and TTS-1-HD models
|
||||
|
||||
## Troubleshooting
|
||||
|
||||
Having issues with TTS? See the [TTS Troubleshooting Guide](troubleshooting.md) for common problems and solutions, including FFmpeg installation and audio playback issues.
|
||||
|
||||
## Enable the Voice agent
|
||||
|
||||
|
||||
54
docs/user-guide/agents/voice/troubleshooting.md
Normal file
54
docs/user-guide/agents/voice/troubleshooting.md
Normal file
@@ -0,0 +1,54 @@
|
||||
# TTS Troubleshooting
|
||||
|
||||
Common issues and solutions for Text-to-Speech functionality in Talemate.
|
||||
|
||||
## FFmpeg Not Found
|
||||
|
||||
Several TTS providers (including [Chatterbox](chatterbox.md) and potentially others) require FFmpeg for audio processing.
|
||||
|
||||
### Symptoms
|
||||
|
||||
You may encounter errors like:
|
||||
|
||||
```
|
||||
Could not load libtorchcodec. Likely causes:
|
||||
1. FFmpeg is not properly installed in your environment. We support
|
||||
versions 4, 5, 6, 7, and 8.
|
||||
2. The PyTorch version is not compatible with this version of TorchCodec.
|
||||
```
|
||||
|
||||
Or:
|
||||
|
||||
```
|
||||
FFmpeg version 8: Could not load this library
|
||||
```
|
||||
|
||||
Or generic FFmpeg-related import/loading errors.
|
||||
|
||||
### Solution
|
||||
|
||||
#### Windows
|
||||
|
||||
Run the included `install-ffmpeg.bat` script from the Talemate root directory:
|
||||
|
||||
```batch
|
||||
install-ffmpeg.bat
|
||||
```
|
||||
|
||||
This will automatically download and install FFmpeg 8.0.1 into your virtual environment.
|
||||
|
||||
#### Linux/macOS
|
||||
|
||||
Install FFmpeg using your system's package manager. FFmpeg versions 4, 5, 6, 7, or 8 are supported.
|
||||
|
||||
### Verification
|
||||
|
||||
After installing FFmpeg, verify it's accessible by running:
|
||||
|
||||
```bash
|
||||
ffmpeg -version
|
||||
```
|
||||
|
||||
You should see output showing the FFmpeg version (4.x, 5.x, 6.x, 7.x, or 8.x are all supported).
|
||||
|
||||
**Important:** Restart Talemate after installing FFmpeg for the changes to take effect.
|
||||
Reference in New Issue
Block a user