Create app.py

app.py

@@ -0,0 +1,61 @@
+# app.py
+from fastapi import FastAPI, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+import torch
+import base64
+import io
+from typing import Dict
+from pydantic import BaseModel
+import numpy as np
+import logging
+from pathlib import Path
+import os
+from tts_service import OptimizedTTSService
+
+# Configure logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+class TTSRequest(BaseModel):
+    text: str
+
+app = FastAPI()
+
+# Add CORS middleware
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# Initialize service with embedded latents
+class TTSServiceConfig:
+    def __init__(self):
+        self.latents_path = "models/goggins_latents.pt"
+        self.model_path = "models/xtts_v2"
+        
+        # Ensure directories exist
+        Path("models").mkdir(exist_ok=True)
+        Path("cache").mkdir(exist_ok=True)
+
+config = TTSServiceConfig()
+service = OptimizedTTSService(config)
+
+@app.post("/generate")
+async def generate_speech(request: TTSRequest):
+    try:
+        logger.info(f"Generating speech for text: {request.text[:50]}...")
+        wav = service.generate_speech(request.text)
+        
+        buffer = io.BytesIO()
+        np.save(buffer, wav.astype(np.float32))
+        
+        return {
+            "status": "success",
+            "audio": base64.b64encode(buffer.getvalue()).decode()
+        }
+    except Exception as e:
+        logger.error(f"Error generating speech: {str(e)}")
+        raise HTTPException(status_code=500, detail=str(e))