mirror of
https://github.com/SpudGunMan/meshing-around.git
synced 2026-10-30 22:01:57 +01:00
68 lines
1.9 KiB
Markdown
68 lines
1.9 KiB
Markdown
# Radio Module: Meshages TTS (Text-to-Speech) Setup
|
||
|
||
The radio module supports audible mesh messages using the [KittenTTS](https://github.com/KittenML/KittenTTS) engine. This allows the bot to generate and play speech from text, making mesh alerts and messages audible on your device.
|
||
|
||
## Features
|
||
|
||
- Converts mesh messages to speech using KittenTTS.
|
||
|
||
## Installation
|
||
|
||
1. **Install Python dependencies:**
|
||
|
||
- `kittentts` is the TTS engine.
|
||
|
||
`pip install https://github.com/KittenML/KittenTTS/releases/download/0.8.1/kittentts-0.8.1-py3-none-any.whl`
|
||
|
||
2. **Install PortAudio (required for sounddevice):**
|
||
|
||
- **macOS:**
|
||
```sh
|
||
brew install portaudio
|
||
```
|
||
- **Linux (Debian/Ubuntu):**
|
||
```sh
|
||
sudo apt-get install portaudio19-dev
|
||
```
|
||
- **Windows:**
|
||
No extra step needed; `sounddevice` will use the default audio driver.
|
||
|
||
## Configuration
|
||
|
||
- Enable TTS in your `config.ini`:
|
||
```ini
|
||
[radioMon]
|
||
meshagesTTS = True
|
||
```
|
||
|
||
## Usage
|
||
|
||
When enabled, the bot will generate and play speech for mesh messages using the selected voice.
|
||
No additional user action is required.
|
||
|
||
## Troubleshooting
|
||
|
||
- If you see errors about missing `sounddevice` or `portaudio`, ensure you have installed the dependencies above.
|
||
- On macOS, you may need to allow microphone/audio access for your terminal.
|
||
- If you have audio issues, check your system’s default output device.
|
||
|
||
## References
|
||
|
||
- [KittenTTS GitHub](https://github.com/KittenML/KittenTTS)
|
||
- [KittenTTS Model on HuggingFace](https://huggingface.co/KittenML/kitten-tts-nano-0.2)
|
||
- [sounddevice documentation](https://python-sounddevice.readthedocs.io/)
|
||
|
||
---
|
||
|
||
# AI Speech with Vosk, Speech to Text
|
||
|
||
"Hey Chirpy"
|
||
|
||
To use VOX detection please install the vosk and sounddevice python modules. Models are around 40MB or less vs wisper
|
||
|
||
- `apt-get install portaudio19-dev`
|
||
- `pip install vosk sounddevice`
|
||
|
||
## References
|
||
|
||
- [Vox Models](https://alphacephei.com/vosk/models) |