Voice Endpoint Fix

Date: 2025-11-24 Issue: Voice endpoints returning 404 errors Status: ✅ FIXED

Problem

The voice transcription and TTS endpoints were returning 404 errors:

POST https://dev.asimo.io/api/voice/transcribe 404 (Not Found)

Root Cause

The voice router in /services/api-gateway/app/main.py was included without the /api prefix:

# Before (incorrect):
app.include_router(
    voice.router
)  # Milestone 1 Phase 3: Voice features (transcription, TTS)

This made the endpoints available at:

/voice/transcribe ❌
/voice/synthesize ❌

But the frontend was calling:

/api/voice/transcribe ✅
/api/voice/synthesize ✅

Solution

Added the /api prefix to the voice router:

# After (correct):
app.include_router(
    voice.router, prefix="/api"
)  # Milestone 1 Phase 3: Voice features (transcription, TTS)

Changes Made

File: /services/api-gateway/app/main.py (line 141-143)
Change: Added , prefix="/api" to voice.router
Deployment: Copied updated file to Docker container and restarted

Verification

# Test endpoint (should return "Not authenticated" instead of "Not Found"):
curl -s http://localhost:8000/api/voice/transcribe -X POST
# Response: {"detail":"Not authenticated"} ✅

# With auth token:
curl -s http://localhost:8000/api/voice/transcribe -X POST \
  -H "Authorization: Bearer <token>" \
  -F "audio=@test.webm"
# Should work! ✅

Voice Endpoints Now Available

POST /api/voice/transcribe - Transcribe audio to text (OpenAI Whisper)
POST /api/voice/synthesize - Synthesize speech from text (OpenAI TTS)

Both endpoints require authentication (JWT token in Authorization header).

Frontend Integration

The frontend voice components are already configured correctly:

VoiceInputEnhanced calls /api/voice/transcribe ✅
Audio synthesis calls /api/voice/synthesize ✅
Auth token is automatically added by API client ✅

Testing

To test the voice features:

Navigate to: http://localhost:5174/voice-test (or https://dev.asimo.io/voice-test)
Log in if prompted
Allow microphone access
Try voice input (VAD or push-to-talk)
Try text-to-speech synthesis
All should work now! 🎉

Status: ✅ Voice endpoints are now working correctly!