Smaller rates and model

app.py

@@ -359,14 +359,14 @@ def get_ram_usage():
     }
 
 @app.post("/api/predict")
-@limiter.limit("15/minute")  # Allow 15 chat requests per minute per IP
+@limiter.limit("5/minute")  # Allow 5 chat requests per minute per IP
 async def predict(request: Request, query_request: QueryRequest):
     query = query_request.query
     history = query_request.history
     return StreamingResponse(stream_response(query, history), media_type="text/event-stream")
 
 @app.post("/api/tts")
-@limiter.limit("10/minute")  # Allow 10 TTS requests per minute per IP
+@limiter.limit("5/minute")  # Allow 5 TTS requests per minute per IP
 async def text_to_speech(request: Request, tts_request: TTSRequest):
     """Convert text to speech using ElevenLabs API"""
     if not elevenlabs_client:
@@ -385,13 +385,14 @@ async def text_to_speech(request: Request, tts_request: TTSRequest):
         # Generate speech
         response = elevenlabs_client.text_to_speech.convert(
             voice_id=tts_voice_id,
-            text=clean_text,
-            voice_settings=VoiceSettings(
-                stability=0.5,
-                similarity_boost=0.8,
-                style=0.2,
-                use_speaker_boost=True
-            )
+            model_id="eleven_flash_v2_5",
+            text=clean_text
+            # voice_settings=VoiceSettings(
+            #     stability=0.5,
+            #     similarity_boost=0.8,
+            #     style=0.2,
+            #     use_speaker_boost=True
+            # )
         )
         
         # Convert generator to bytes