Update LLMwithvoice.py

LLMwithvoice.py

@@ -4,7 +4,7 @@ import numpy as np
 from transformers import AutoTokenizer, AutoModelForQuestionAnswering
 from parler_tts import ParlerTTSForConditionalGeneration
 from pydub import AudioSegment
-import simpleaudio as sa
+from IPython.display import Audio, display
 
 # Hugging Face API URL for Roberta model
 API_URL_ROBERTA = "https://api-inference.huggingface.co/models/deepset/roberta-base-squad2"
@@ -54,17 +54,12 @@ def generate_speech(answer):
         channels=1
     )
 
-    # Play the audio using simpleaudio
-    try:
-        play_obj = sa.play_buffer(
-            audio_segment.raw_data,
-            num_channels=1,
-            bytes_per_sample=audio_segment.sample_width,
-            sample_rate=audio_segment.frame_rate
-        )
-        play_obj.wait_done()  # Wait until the audio is done playing
-    except Exception as e:
-        print(f"Error playing audio: {e}")
+    # Save the audio to a file
+    audio_file = "output.wav"
+    audio_segment.export(audio_file, format="wav")
+
+    # Play the audio using IPython.display.Audio
+    display(Audio(audio_file))
 
 # Function to interface with Gradio
 def gradio_interface(api_token, prompt, context):
@@ -72,13 +67,4 @@ def gradio_interface(api_token, prompt, context):
     if 'error' in answer:
         return answer['error'], None
     generate_speech(answer.get('answer', ''))
-    return answer.get('answer', 'No answer found'), None
-
-# Example usage
-if __name__ == "__main__":
-    api_token = "your_huggingface_api_token"
-    prompt = "What is the capital of France?"
-    context = "France, in Western Europe, encompasses medieval cities, alpine villages, and Mediterranean beaches. Paris, its capital, is famed for its fashion houses, classical art museums including the Louvre, and monuments like the Eiffel Tower."
-
-    answer, _ = gradio_interface(api_token, prompt, context)
-    print("Answer:", answer)
+    return answer.get('answer', 'No answer found'), None