change to cpu

app.py

@@ -172,18 +172,11 @@ with gr.Blocks(title="Voice chat with LLM") as demo:
         def handle_speech_generation(sentence, chatbot_history, chatbot_voice):
             if sentence != "":
                 print("Processing sentence")
-                # generate speech by cloning a voice using default settings
-                xtts_model.tts_to_file(text="It took me quite a long time to develop a voice, and now that I have it I'm not going to be silent.",
-                                file_path="output.wav",
-                                speaker_wav=[f"examples/{chatbot_voice}.wav"],
-                                language="en",
-                                split_sentences=True
-                                )
-                yield (sentence, chatbot_history, wave_header_chunk())
-                #generated_speech = generate_speech_for_sentence(chatbot_history, chatbot_voice, sentence, xtts_model, None, return_as_byte=True)
-                #if generated_speech is not None:
-                #    _, audio_dict = generated_speech
-                #    yield (sentence, chatbot_history, audio_dict["value"])
+                # generate speech by cloning a voice using default setting
+                generated_speech = generate_speech_for_sentence(chatbot_history, chatbot_voice, sentence, xtts_model, None, return_as_byte=True)
+                if generated_speech is not None:
+                    _, audio_dict = generated_speech
+                    yield (sentence, chatbot_history, audio_dict["value"])
 
         if initial_greeting:
             # Process only the initial greeting if specified