Update app.py

app.py

@@ -1,33 +1,45 @@
 import gradio as gr
 from transformers import pipeline
+from gtts import gTTS
+import os
 
 # Initialize NLP pipelines
 qa = pipeline("question-answering", model="distilbert-base-uncased-distilled-squad")
-summarizer = pipeline("summarization", model="t5-small")
-tts = pipeline("text-to-speech", model="facebook/mms-tts-eng")
+try:
+    summarizer = pipeline("summarization", model="t5-small")
+except Exception as e:
+    print(f"Error loading summarizer: {e}")
+    summarizer = None
 
 def study_aid(question, context, font_size=16, audio_output=False, simplify_text=False):
-    # Log decisions for transparency
     with open("decision_log.txt", "a") as f:
         f.write(f"Question: {question}, Simplified: {simplify_text}, Audio: {audio_output}, Font: {font_size}\n")
     
-    # Simplify context if requested
-    if simplify_text:
-        context = summarizer(context, max_length=100, min_length=50)[0]["summary_text"]
+    simplified_context = context
+    if simplify_text and summarizer is not None:
+        try:
+            # Ensure input length is suitable
+            if len(context.split()) < 10:
+                simplified_context = "Input too short to simplify."
+            elif len(context.split()) > 512:
+                simplified_context = "Input too long to simplify."
+            else:
+                summary = summarizer(context, max_length=100, min_length=50)[0]["summary_text"]
+                simplified_context = summary
+        except Exception as e:
+            simplified_context = f"Error simplifying text: {str(e)}"
     
-    # Answer the question
-    answer = qa(question=question, context=context)["answer"]
+    answer = qa(question=question, context=simplified_context)["answer"]
     
-    # Format output with adjustable font size
     output = f"<div style='font-size:{font_size}px'>"
-    if simplify_text:
-        output += f"<b>Simplified Context:</b> {context}<br>"
+    if simplify_text and simplified_context != context:
+        output += f"<b>Simplified Context:</b> {simplified_context}<br>"
     output += f"<b>Answer:</b> {answer}</div>"
     
-    # Generate audio if requested
     if audio_output:
-        audio = tts(answer)
-        return output, audio
+        tts = gTTS(text=answer, lang='en')
+        tts.save("answer_audio.mp3")
+        return output, "answer_audio.mp3"
     
     return output, None
 
@@ -36,7 +48,6 @@ def submit_feedback(feedback):
         f.write(feedback + "\n")
     return "Feedback submitted!"
 
-# Create Gradio app with Blocks
 with gr.Blocks(title="StudyBuddy: Accessible Study Aid for Neurodiverse Students") as app:
     gr.Markdown(
         """
@@ -45,7 +56,6 @@ with gr.Blocks(title="StudyBuddy: Accessible Study Aid for Neurodiverse Students
         """
     )
     
-    # Tab 1: Study Aid
     with gr.Tab("Ask a Question"):
         question_input = gr.Textbox(label="Question", placeholder="e.g., What is machine learning?")
         context_input = gr.Textbox(label="Context (Lecture Notes)", placeholder="Paste your notes here...")
@@ -62,7 +72,6 @@ with gr.Blocks(title="StudyBuddy: Accessible Study Aid for Neurodiverse Students
             outputs=[study_output_text, study_output_audio]
         )
     
-    # Tab 2: Feedback
     with gr.Tab("Submit Feedback"):
         feedback_input = gr.Textbox(label="Feedback", placeholder="Report issues or suggestions...")
         feedback_submit_btn = gr.Button("Submit Feedback")
@@ -74,4 +83,7 @@ with gr.Blocks(title="StudyBuddy: Accessible Study Aid for Neurodiverse Students
             outputs=feedback_output
         )
 
+    with gr.Tab("View Logs"):
+        logs_output = gr.Textbox(label="Decision Logs", value=lambda: open("decision_log.txt", "r").read())
+
 app.launch()