Update app.py

app.py

@@ -7,6 +7,7 @@ from fastapi.security import HTTPBasic, HTTPBasicCredentials
 from fastapi.responses import FileResponse
 import nemo.collections.asr as nemo_asr
 from piper.voice import PiperVoice
+from pydantic import BaseModel
 
 app = FastAPI(title="ASR & TTS API")
 security = HTTPBasic()
@@ -78,8 +79,11 @@ async def transcribe(file: UploadFile = File(...), _: str = Depends(verify_crede
     os.remove(audio_path)
     return {"text": transcription}
 
+class TTSRequest(BaseModel):
+    text: str
+    
 @app.post("/tts")
-async def synthesize(text: str, _: str = Depends(verify_credentials)):
+async def synthesize(req: TTSRequest, _: str = Depends(verify_credentials)):
     if not tts_voice:
         raise HTTPException(status_code=503, detail="TTS model not loaded")
     
@@ -93,8 +97,14 @@ async def synthesize(text: str, _: str = Depends(verify_credentials)):
             wav_file.setnchannels(1)
             wav_file.setsampwidth(2)
             wav_file.setframerate(tts_voice.config.sample_rate)
-            tts_voice.synthesize(text, wav_file)
+            # Make sure to call req.text here!
+            tts_voice.synthesize(req.text, wav_file)
 
     await loop.run_in_executor(None, generate_audio)
         
-    return FileResponse(output_path, media_type="audio/wav")
+    return FileResponse(
+        output_path, 
+        media_type="audio/wav",
+        headers={"Content-Disposition": "attachment; filename=tts_output.wav"}
+    )
+