# Discord Voice Bot (GLaDOS)

## Overview
Voice-controlled Discord bot with conversational AI — always available in #coding voice channel

---

## Location
`/home/admin/.openclaw/workspace/discord-voice-bot/`

---

## Architecture (Completed 2026-02-08)

### Components
| Component | Endpoint | Purpose |
|-----------|----------|---------|
| **TTS** | 192.168.0.17:5050 | GLaDOS voice synthesis |
| **STT** | 192.168.0.17:10300 | Whisper via Wyoming protocol |
| **LLM** | 192.168.0.17:11434 | Ollama (qwen3-coder-next:cloud) |
| **Auto-join** | On startup | Joins #coding voice channel |

### Voice Pipeline
```
Wake Word → STT (Whisper) → LLM (Ollama) → TTS (GLaDOS) → Voice Output
```

---

## Commands
- `!join` — Join voice channel
- `!leave` — Leave voice channel
- `!test [text]` — Test TTS synthesis
- `!listen` — 5-second voice recording (7-10 sec latency)

---

## MCP Integration (OpenClaw Bridge)

### MCP Tools Available via Voice
- `list_projects()` — "What am I working on?"
- `create_project(name)` — Start new project
- `add_task(project, task)` — Add tasks via voice
- `update_task_status(taskId, status)` — Mark complete
- `get_project_status()` — Check overall progress

### MCP Server Config
```yaml
mcp_servers:
  - name: "openclaw"
    transport: "stdio"
    command: "python"
    args:
      - "C:\Users\admin\.openclaw\workspace\discord-voice-bot\openclaw_mcp_server.py"
```

---

## Files Created
- `openclaw_mcp_server.py` — FastMCP server exposing OpenClaw tools
- `main.py` — Bot implementation using discord.py, wyoming
- `config.yaml` — Configuration
- `requirements.txt` — Dependencies (discord.py, requests, numpy, wyoming)

---

## What Changed (v2)
- Removed GLaDOS direct dependencies (simpler integration)
- Uses existing services (Whisper, Ollama, HTTP TTS)
- Auto-joins voice channel on startup
- Clean async architecture
- MCP bridge for OpenClaw integration

---

## Known Limitations
- Voice latency: 7-10 seconds for `!listen` (5s recording + processing)
- Not conversational due to latency
- Best for command-based interactions, not real-time chat

---

## Status
✅ **COMPLETED** — Operational in #coding voice channel

Bot auto-joins on startup and accepts voice commands via MCP bridge.

---

*Status: Complete | Version: 2.0 | Updated: 2026-02-22*