Update app.py

app.py

@@ -1,21 +1,14 @@
 import streamlit as st
-import outetts
-from scipy.io.wavfile import write
+from TTS.api import TTS
 import tempfile
 import os
 from pydub import AudioSegment
 
-# Initialize model configuration
-model_config = outetts.HFModelConfig_v1(
-    model_path="OuteAI/OuteTTS-0.2-500M",
-    language="en"  # Supported languages: en, zh, ja, ko
-)
-
-# Initialize the interface
-interface = outetts.InterfaceHF(model_version="0.2", cfg=model_config)
+# Initialize the TTS model
+tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2", gpu=True)
 
 # Streamlit UI
-st.title("OuteTTS Speech Synthesis")
+st.title("XTTS v2 Speech Synthesis")
 st.write("Enter text below to generate speech.")
 
 # Sidebar for reference voice
@@ -38,18 +31,17 @@ text_input = st.text_area("Text to convert to speech:", "Hello, this is an AI-ge
 
 if st.button("Generate Speech"):
     with st.spinner("Generating audio..."):
-        # Generate speech with reference audio
-        output = interface.generate(
+        # Define output path
+        output_path = "output.wav"
+        
+        # Generate speech using XTTS v2
+        tts.tts_to_file(
             text=text_input,
-            temperature=0.1,
-            repetition_penalty=1.1,
-            max_length=4096
+            file_path=output_path,
+            speaker_wav=ref_audio_path if ref_audio_path else None,
+            language="en"
         )
         
-        # Save the synthesized speech to a file
-        output_path = "output.wav"
-        output.save(output_path)
-        
         # Play the audio in the Streamlit app
         st.audio(output_path, format="audio/wav")
         st.success("Speech generated successfully!")