Update app.py

app.py

@@ -24,7 +24,7 @@ def audio_to_text(audio):
     outputs = whisper(**inputs)
     predicted_ids = torch.argmax(outputs.logits, dim=-1)
     transcribed_text = whisper_tokenizer.decode(predicted_ids[0])
-    return str(transcribed_text)
+    return transcribed_text
 
 # Define functions for model predictions
 def generate_text(prompt):
@@ -36,11 +36,14 @@ def text_to_speech(text):
     speech = vits_tts(generated_text)[0]['audio']
     return speech
 
+def my_alexa(audio):
+    return text_to_speech(generate_text(audio_to_text(audio)))
+
 # Create Gradio interface
 audio_input = gr.Audio(label="Can I Help You?", sources="microphone")
 text_output = gr.Textbox(label="Transcribed Text")
 gr.Interface(
-    text_to_speech, 
+    my_alexa, 
     inputs=audio_input, 
     outputs=gr.Audio(label="Answer:"),
     title="MyAlexa AI Assistant").launch(inbrowser=True, debug=True)