feat: VoiceCoach v2.1 - GPU local, Ollama LLM, settings panel, token server separation

- pipeline.py: switched LLM from OpenClaw/Claude to local Ollama, added settings loading, sentence aggregator, async SpeechBrain, tuned VAD - main.py: simplified (no more token server thread, just pipeline) - token_server.py: NEW - standalone token + settings API server - docker-compose.yml: all services on GPU machine, added token-server service, CUDA speaches, Ollama LLM - web-client/index.html: added settings panel with sliders - web-client/app.js: settings management, localStorage persistence, push to server - web-client/style.css: settings panel styles - web-client/nginx.conf: proxy to token-server instead of pipecat-agent - patch_livekit.py: updated with proper indentation-safe patching
2026-04-04 12:27:06 -05:00
parent f62772663d
commit d901208032
12 changed files with 923 additions and 229 deletions
@@ -0,0 +1,95 @@
+"""
+Standalone LiveKit token server + settings API.
+Runs independently from the pipeline so it never goes down when the pipeline crashes.
+Settings are stored in a JSON file so the pipeline can read them on startup.
+"""
+import json
+import os
+from pathlib import Path
+import uvicorn
+from fastapi import FastAPI
+from pydantic import BaseModel
+from typing import Optional
+from livekit import api as livekit_api
+
+LIVEKIT_API_KEY = os.environ["LIVEKIT_API_KEY"]
+LIVEKIT_API_SECRET = os.environ["LIVEKIT_API_SECRET"]
+LIVEKIT_PUBLIC_URL = os.getenv("LIVEKIT_PUBLIC_URL", "wss://voice.hobbs.farm")
+SETTINGS_PATH = Path(os.getenv("SETTINGS_PATH", "/data/settings.json"))
+
+app = FastAPI(title="VoiceCoach Token + Settings Service")
+
+DEFAULTS = {
+    "confidence": 0.7,
+    "min_volume": 0.3,
+    "start_secs": 0.4,
+    "stop_secs": 0.5,
+    "max_tokens": 150,
+    "tts_voice": "af_nova",
+}
+
+
+def load_settings() -> dict:
+    try:
+        if SETTINGS_PATH.exists():
+            return {**DEFAULTS, **json.loads(SETTINGS_PATH.read_text())}
+    except Exception:
+        pass
+    return {**DEFAULTS}
+
+
+def save_settings(settings: dict):
+    SETTINGS_PATH.parent.mkdir(parents=True, exist_ok=True)
+    SETTINGS_PATH.write_text(json.dumps(settings, indent=2))
+
+
+class TokenRequest(BaseModel):
+    identity: str = "jeremy"
+
+
+class TokenResponse(BaseModel):
+    token: str
+    url: str
+
+
+class Settings(BaseModel):
+    confidence: Optional[float] = None
+    min_volume: Optional[float] = None
+    start_secs: Optional[float] = None
+    stop_secs: Optional[float] = None
+    max_tokens: Optional[int] = None
+    tts_voice: Optional[str] = None
+
+
+@app.get("/health")
+async def health():
+    return {"status": "ok", "version": "2.1"}
+
+
+@app.post("/token")
+async def get_token(req: TokenRequest):
+    token = livekit_api.AccessToken(LIVEKIT_API_KEY, LIVEKIT_API_SECRET)
+    token.with_grants(livekit_api.VideoGrants(
+        room_join=True, room="voicecoach",
+        can_publish=True, can_subscribe=True,
+    ))
+    token.identity = req.identity
+    return TokenResponse(token=token.to_jwt(), url=LIVEKIT_PUBLIC_URL)
+
+
+@app.get("/settings")
+async def get_settings():
+    return load_settings()
+
+
+@app.post("/settings")
+async def update_settings(s: Settings):
+    current = load_settings()
+    update = {k: v for k, v in s.dict().items() if v is not None}
+    current.update(update)
+    save_settings(current)
+    return current
+
+
+if __name__ == "__main__":
+    uvicorn.run(app, host="0.0.0.0", port=8080, log_level="info")