Update app.py

app.py

@@ -13,6 +13,7 @@ from pydub import AudioSegment
 from dotenv import load_dotenv
 import tempfile
 import spaces
+from TTS.api import TTS
 
 # Load environment variables
 load_dotenv()
@@ -55,18 +56,15 @@ def generate_script(user_prompt: str, model_id: str, token: str, duration: int):
 # Voice-Over Generation Function
 # ---------------------------------------------------------------------
 @spaces.GPU(duration=300)
-def generate_voice(script: str, speaker: str):
+def generate_voice(script: str, speaker: str = "default"):
     try:
-        # Replace with your chosen TTS model
-        tts_model = "coqui/XTTS-v2"
-        processor = AutoProcessor.from_pretrained(tts_model)
-        model = AutoModelForCausalLM.from_pretrained(tts_model)
-
-        inputs = processor(script, return_tensors="pt")
-        speech = model.generate(**inputs)
-
+        # Load the TTS model
+        tts = TTS(model_name="tts_models/en/ljspeech/tacotron2-DDC", gpu=torch.cuda.is_available())
+        
+        # Generate the speech audio file
         output_path = f"{tempfile.gettempdir()}/generated_voice.wav"
-        write(output_path, 22050, speech.cpu().numpy())
+        tts.tts_to_file(text=script, file_path=output_path, speaker=speaker)
+        
         return output_path
     except Exception as e:
         return f"Error generating voice-over: {e}"