Update main.py

main.py

@@ -1,6 +1,6 @@
 """
 Production FastAPI Server for Public Speaking Coach
-Handles audio uploads and returns comprehensive analysis
+With LLM Tips and Avatar Voice Support
 """
 
 import os
@@ -12,28 +12,36 @@ from typing import Optional
 import uvicorn
 from fastapi import FastAPI, UploadFile, File, HTTPException, status
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import JSONResponse
+from fastapi.responses import JSONResponse, FileResponse
+from fastapi.staticfiles import StaticFiles
 from pydantic import BaseModel
 
-from kid_coach_pipeline import EnhancedPublicSpeakingCoach
+from enhanced_speech_coach_with_llm import EnhancedPublicSpeakingCoach
 
 # ================= APP CONFIGURATION =================
 
 app = FastAPI(
     title="Public Speaking Coach API",
-    description="AI-powered speech analysis and coaching for all ages",
-    version="2.0.0"
+    description="AI-powered speech analysis with LLM tips and avatar voice",
+    version="3.0.0"
 )
 
-# CORS Configuration - Adjust for production
+# CORS Configuration
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["*"],  # Change to specific domains in production
+    allow_origins=["*"],
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
 )
 
+# Audio output directory
+AUDIO_OUTPUT_DIR = "/tmp/audio_outputs"
+os.makedirs(AUDIO_OUTPUT_DIR, exist_ok=True)
+
+# Mount static files for audio serving
+app.mount("/audio", StaticFiles(directory=AUDIO_OUTPUT_DIR), name="audio")
+
 # Global engine instance
 coach_engine: Optional[EnhancedPublicSpeakingCoach] = None
 
@@ -53,6 +61,7 @@ class HealthResponse(BaseModel):
     """Health check response"""
     status: str
     engine_loaded: bool
+    tts_enabled: bool
     supported_formats: list
 
 
@@ -75,7 +84,10 @@ async def startup_event():
     
     try:
         print("\n📦 Loading AI models...")
-        coach_engine = EnhancedPublicSpeakingCoach(whisper_model_size="base")
+        coach_engine = EnhancedPublicSpeakingCoach(
+            whisper_model_size="base",
+            enable_tts=True
+        )
         print("✅ Coach engine ready!")
         print("\n" + "="*60)
         print("🎤 API is ready to analyze speeches!")
@@ -101,6 +113,7 @@ async def root():
     return {
         "status": "online",
         "engine_loaded": coach_engine is not None,
+        "tts_enabled": coach_engine.tts_enabled if coach_engine else False,
         "supported_formats": list(SUPPORTED_FORMATS)
     }
 
@@ -111,6 +124,7 @@ async def health_check():
     return {
         "status": "healthy" if coach_engine else "degraded",
         "engine_loaded": coach_engine is not None,
+        "tts_enabled": coach_engine.tts_enabled if coach_engine else False,
         "supported_formats": list(SUPPORTED_FORMATS)
     }
 
@@ -118,19 +132,13 @@ async def health_check():
 @app.post("/coach")
 async def analyze_speech(file: UploadFile = File(...)):
     """
-    Main endpoint: Upload audio file and receive speech analysis
+    Main endpoint: Upload audio file and receive comprehensive analysis
     
-    Args:
-        file: Audio file (wav, mp3, m4a, flac, ogg, etc.)
-        
     Returns:
-        Comprehensive speech analysis with scores and coaching feedback
-        
-    Raises:
-        400: Invalid file format or corrupted audio
-        413: File too large
-        500: Analysis failed
-        503: Engine not loaded
+    - Full speech analysis
+    - Personalized tips from LLM
+    - Improved transcript
+    - Avatar voice audio (if TTS enabled)
     """
     
     # Check if engine is loaded
@@ -189,7 +197,7 @@ async def analyze_speech(file: UploadFile = File(...)):
         print(f"\n📁 Processing: {file.filename} ({len(content) / 1024:.1f} KB)")
         
         # Run analysis
-        result = coach_engine.analyze_speech(temp_file)
+        result = coach_engine.analyze_speech(temp_file, output_dir=AUDIO_OUTPUT_DIR)
         
         # Check for analysis errors
         if "error" in result:
@@ -232,6 +240,24 @@ async def analyze_speech_alias(file: UploadFile = File(...)):
     return await analyze_speech(file)
 
 
+@app.get("/audio/{filename}")
+async def get_audio(filename: str):
+    """Serve generated avatar audio files"""
+    file_path = os.path.join(AUDIO_OUTPUT_DIR, filename)
+    
+    if not os.path.exists(file_path):
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Audio file not found"
+        )
+    
+    return FileResponse(
+        file_path,
+        media_type="audio/wav",
+        filename=filename
+    )
+
+
 # ================= ERROR HANDLERS =================
 
 @app.exception_handler(HTTPException)
@@ -268,6 +294,6 @@ if __name__ == "__main__":
     uvicorn.run(
         app,
         host="0.0.0.0",
-        port=7860,
+        port=8000,
         log_level="info"
     )