test

app.py

@@ -20,21 +20,21 @@ from melo.api import TTS
 speed = 1.0
 device = 'cpu' # or cuda:0
 
-model = TTS(language='EN', device=device)
-speaker_ids = model.hps.data.spk2id
-
 # Pydantic model for request body
 class TextInput(BaseModel):
     text: str
+    lang: str
+    speaker: str
 
 @app.post("/generate-tts")
 async def generate_tts(input: TextInput):
     try:
+        model = TTS(language=input.lang, device=device)
+        speaker_ids = model.hps.data.spk2id
         # Generate audio to a temporary buffer
         buffer = io.BytesIO()
-        model.tts_to_file(input.text, speaker_ids['EN-US'], buffer, speed=speed, format='wav')
+        model.tts_to_file(input.text, speaker_ids[input.speaker], buffer, speed=speed, format='wav')
         buffer.seek(0)
-        
         # Return audio as streaming response
         return StreamingResponse(
             buffer,