Update app.py

app.py

@@ -1,6 +1,5 @@
 import gradio as gr
 import torch
-import speech_recognition as sr
 from groq import Groq
 import os
 import tempfile
@@ -45,22 +44,6 @@ def predict_text_emotion(text):
         logger.error(f"Error with Groq API (text emotion): {str(e)}")
         return "neutral"
 
-def transcribe_audio(audio_path):
-    r = sr.Recognizer()
-    try:
-        with sr.AudioFile(audio_path) as source:
-            audio_text = r.listen(source)
-        text = r.recognize_google(audio_text)
-        return text
-    except sr.UnknownValueError:
-        return "I didn’t catch that—could you try again?"
-    except sr.RequestError as e:
-        logger.error(f"Speech recognition error: {str(e)}")
-        return "Speech recognition unavailable—try typing instead."
-    except Exception as e:
-        logger.error(f"Unexpected error in audio transcription: {str(e)}")
-        return "Error processing audio."
-
 def generate_response(user_input, emotion):
     prompt = f"The user is feeling {emotion}. They said: '{user_input}'. Respond in a friendly caring manner with the user so the user feels being loved."
     try:
@@ -87,14 +70,12 @@ def text_to_speech(text):
         logger.error(f"Error generating speech: {str(e)}")
         return None
 
-# Chat function for Gradio with voice output
+# Chat function for Gradio with voice output (text input only)
 def chat_function(input_type, text_input, audio_input, chat_history):
     if input_type == "text" and text_input:
         user_input = text_input
-    elif input_type == "voice" and audio_input:
-        user_input = transcribe_audio(audio_input)
     else:
-        return chat_history, "Please provide text or voice input.", gr.update(value=text_input), None
+        return chat_history, "Please provide text input. Voice input is not supported.", gr.update(value=text_input), None
 
     emotion = predict_text_emotion(user_input)
     response = generate_response(user_input, emotion)
@@ -129,48 +110,53 @@ css = """
 """
 
 # Build the Gradio interface
-with gr.Blocks(theme=gr.themes.Soft(), css=css) as app:
-    gr.Markdown(
-        """
-        # Multimodal Mental Health AI Agent
-        Chat with our empathetic AI designed to support you by understanding your emotions through text and voice.
-        """
-    )
-
-    with gr.Row():
-        with gr.Column(scale=1):
-            emotion_display = gr.Textbox(label="Emotion", interactive=False, placeholder="Detected emotion will appear here")
-        
-        with gr.Column(scale=3):
-            chatbot = gr.Chatbot(label="Conversation History", height=500, type="messages", elem_classes="chatbot")
-
-    with gr.Row(elem_classes="input-container"):
-        input_type = gr.Radio(["text", "voice"], label="Input Method", value="text")
-        text_input = gr.Textbox(label="Type Your Message", placeholder="How are you feeling today?", visible=True)
-        audio_input = gr.Audio(sources=["microphone"], type="filepath", label="Record Your Message", visible=False)
-        submit_btn = gr.Button("Send", variant="primary")
-        clear_btn = gr.Button("Clear Chat", variant="secondary")
-        audio_output = gr.Audio(label="Assistant Response", type="filepath", interactive=False, autoplay=True)
-
-    # Dynamic visibility based on input type
-    def update_visibility(input_type):
-        return gr.update(visible=input_type == "text"), gr.update(visible=input_type == "voice")
-
-    input_type.change(fn=update_visibility, inputs=input_type, outputs=[text_input, audio_input])
-
-    # Submit action with voice output
-    submit_btn.click(
-        fn=chat_function,
-        inputs=[input_type, text_input, audio_input, chatbot],
-        outputs=[chatbot, emotion_display, text_input, audio_output]
-    )
-
-    # Clear chat and audio
-    clear_btn.click(
-        lambda: ([], "", "", None),
-        inputs=None,
-        outputs=[chatbot, emotion_display, text_input, audio_output]
-    )
+try:
+    with gr.Blocks(theme=gr.themes.Soft(), css=css) as app:
+        gr.Markdown(
+            """
+            # Multimodal Mental Health AI Agent
+            Chat with our empathetic AI designed to support you by understanding your emotions through text.
+            """
+        )
+
+        with gr.Row():
+            with gr.Column(scale=1):
+                emotion_display = gr.Textbox(label="Emotion", interactive=False, placeholder="Detected emotion will appear here")
+            
+            with gr.Column(scale=3):
+                chatbot = gr.Chatbot(label="Conversation History", height=500, type="messages", elem_classes="chatbot")
+
+        with gr.Row(elem_classes="input-container"):
+            input_type = gr.Radio(["text", "voice"], label="Input Method", value="text")
+            text_input = gr.Textbox(label="Type Your Message", placeholder="How are you feeling today?", visible=True)
+            audio_input = gr.Audio(sources=["microphone"], type="filepath", label="Record Your Message", visible=False)
+            submit_btn = gr.Button("Send", variant="primary")
+            clear_btn = gr.Button("Clear Chat", variant="secondary")
+            audio_output = gr.Audio(label="Assistant Response", type="filepath", interactive=False, autoplay=True)
+
+        # Dynamic visibility based on input type
+        def update_visibility(input_type):
+            return gr.update(visible=input_type == "text"), gr.update(visible=input_type == "voice")
+
+        input_type.change(fn=update_visibility, inputs=input_type, outputs=[text_input, audio_input])
+
+        # Submit action with voice output
+        submit_btn.click(
+            fn=chat_function,
+            inputs=[input_type, text_input, audio_input, chatbot],
+            outputs=[chatbot, emotion_display, text_input, audio_output]
+        )
+
+        # Clear chat and audio
+        clear_btn.click(
+            lambda: ([], "", "", None),
+            inputs=None,
+            outputs=[chatbot, emotion_display, text_input, audio_output]
+        )
+
+except Exception as e:
+    logger.error(f"Error initializing Gradio interface: {str(e)}")
+    raise
 
 # Launch the app (commented out for Hugging Face Spaces)
 # if __name__ == "__main__":