Changelog
[1.0.0] — 2026-05-16
First stable release.
Features
- Slash commands:
/ask,/say,/askvoice,/listen,/stoplisten,/language,/synthesizer,/voice,/settings,/reset,/leave - LLM backends: Ollama (default), Groq, OpenAI, custom OpenAI-compatible APIs
- Voice listen: hybrid Google + Whisper STT, optional Edge TTS replies in voice
- Per-user language and TTS settings (11 languages)
- NVIDIA GPU Whisper (
cuda+float16) with CPU fallback - Local logs in
data/logs/bot.log
Requirements
- Python 3.10+, FFmpeg, Discord bot token
- For GPU STT on Windows:
.\scripts\install-gpu-stt.ps1