Update app.py

app.py

@@ -1,11 +1,11 @@
-import os
 import gradio as gr
 import openai
 import speech_recognition as sr
 import logging
+import traceback
 
 # Set up logging
-logging.basicConfig(level=logging.INFO)
+logging.basicConfig(level=logging.DEBUG)
 logger = logging.getLogger(__name__)
 
 # Set OpenAI API key
@@ -27,6 +27,8 @@ def generate_text():
 
 def get_pronunciation_feedback(original_text, transcription):
     try:
+        logger.debug(f"Original text: {original_text}")
+        logger.debug(f"Transcription: {transcription}")
         response = openai.ChatCompletion.create(
             model="gpt-3.5-turbo",
             messages=[
@@ -34,73 +36,52 @@ def get_pronunciation_feedback(original_text, transcription):
                 {"role": "user", "content": f"Original text: '{original_text}'\nTranscription: '{transcription}'\nProvide pronunciation feedback."}
             ]
         )
-        return response.choices[0].message['content']
+        feedback = response.choices[0].message['content']
+        logger.debug(f"Generated feedback: {feedback}")
+        return feedback
     except Exception as e:
         logger.error(f"Error in get_pronunciation_feedback: {str(e)}")
+        logger.error(traceback.format_exc())
         return "Error generating feedback. Please try again."
 
 def transcribe_audio_realtime(audio):
+
+
+
     try:
+        logger.debug(f"Received audio file: {audio}")
         recognizer = sr.Recognizer()
         with sr.AudioFile(audio) as source:
+            logger.debug("Reading audio file")
             audio_data = recognizer.record(source)
-        return recognizer.recognize_google(audio_data)
+        logger.debug("Transcribing audio")
+        transcription = recognizer.recognize_google(audio_data)
+        logger.debug(f"Transcription result: {transcription}")
+        return transcription
     except sr.UnknownValueError:
+        logger.warning("Could not understand audio")
         return "Could not understand audio"
     except sr.RequestError as e:
         logger.error(f"Could not request results from the speech recognition service; {str(e)}")
         return "Error in speech recognition service"
     except Exception as e:
         logger.error(f"Error in transcribe_audio_realtime: {str(e)}")
+        logger.error(traceback.format_exc())
         return "Error transcribing audio. Please try again."
 
 def practice_pronunciation(audio, text_to_read):
+    logger.info("Starting practice_pronunciation function")
     if not text_to_read:
+        logger.info("Generating new text to read")
         text_to_read = generate_text()
-    transcription = transcribe_audio_realtime(audio)
-    feedback = get_pronunciation_feedback(text_to_read, transcription)
-    return text_to_read, transcription, feedback
-
-# Custom CSS for improved styling
-custom_css = """
-.container {max-width: 800px; margin: auto; padding: 20px;}
-.title {text-align: center; color: #2c3e50; margin-bottom: 20px;}
-.subtitle {text-align: center; color: #34495e; margin-bottom: 30px;}
-.input-section, .output-section {background-color: #ecf0f1; padding: 20px; border-radius: 10px; margin-bottom: 20px;}
-.input-section h3, .output-section h3 {color: #2980b9; margin-bottom: 10px;}
-.button-primary {background-color: #3498db !important;}
-.button-secondary {background-color: #2ecc71 !important;}
-"""
-
-# Gradio interface with improved UI
-with gr.Blocks(css=custom_css) as demo:
-    gr.HTML("<div class='container'>")
-    gr.HTML("<h1 class='title'>Pronunciation Practice Tool</h1>")
-    gr.HTML("<p class='subtitle'>Improve your English pronunciation with AI-powered feedback</p>")
+    logger.info(f"Text to read: {text_to_read}")
     
-    with gr.Box(className="input-section"):
-        gr.HTML("<h3>Step 1: Get Text to Read</h3>")
-        with gr.Row():
-            text_to_read = gr.Textbox(label="Text to Read", placeholder="Click 'Generate New Text' or type your own text here")
-            generate_button = gr.Button("Generate New Text", variant="primary", className="button-primary")
-    
-    with gr.Box(className="input-section"):
-        gr.HTML("<h3>Step 2: Record Your Voice</h3>")
-        audio_input = gr.Audio(type="filepath", label="Record your voice reading the text above")
-    
-    with gr.Box(className="output-section"):
-        gr.HTML("<h3>Step 3: Get Feedback</h3>")
-        with gr.Row():
-            transcription_output = gr.Textbox(label="Your Transcription", lines=3)
-            feedback_output = gr.Textbox(label="Pronunciation Feedback", lines=5)
+    logger.info("Starting transcription")
+    transcription = transcribe_audio_realtime(audio)
+    logger.info(f"Transcription result: {transcription}")
     
-    submit_button = gr.Button("Submit for Feedback", variant="secondary", className="button-secondary")
+    logger.info("Getting pronunciation feedback")
+    feedback = get_pronunciation_feedback(text_to_read, transcription)
+    logger.info(f"Feedback generated: {feedback}")
     
-    generate_button.click(generate_text, outputs=text_to_read)
-    submit_button.click(practice_pronunciation, inputs=[audio_input, text_to_read], outputs=[text_to_read, transcription_output, feedback_output])
-
-gr.HTML("</div>")
-
-# Launch the app
-if __name__ == "__main__":
-    demo.launch()
+    return text_to_read, transcription, feedback