Update app.py

app.py

@@ -1,10 +1,12 @@
 import gradio as gr
-from transformers import pipeline, TFAutoModelForSeq2SeqLM, AutoTokenizer
+from transformers import pipeline
 import torch
 
 # Initialize Hugging Face pipelines
 speech_to_text = pipeline("automatic-speech-recognition", model="openai/whisper-large")
-text_to_speech = pipeline("text-to-speech", model="facebook/tacotron2", device=0)  # Set device to CPU (0) or GPU (cuda)
+
+# Use a valid TTS model for text-to-speech (VITS model from Hugging Face)
+text_to_speech = pipeline("text-to-speech", model="espnet/kan-bayashi_ljspeech_vits", device=0)  # Use CPU or GPU (cuda)
 
 # Function to process speech to text and text to speech
 def process_audio(input_audio):
@@ -12,7 +14,7 @@ def process_audio(input_audio):
     recognized_text = speech_to_text(input_audio)["text"]
     print(f"Recognized text: {recognized_text}")
     
-    # Process the text to speech using Tacotron2 model
+    # Process the text to speech using the TTS model
     audio_response = text_to_speech(recognized_text)
     return audio_response, recognized_text