Update app.py

app.py

@@ -6,9 +6,6 @@ from google import genai
 from google.genai import types
 import asyncio
 import concurrent.futures
-from streamlit_audiorecorder import st_audiorecorder
-import pydub
-
 
 GEMINI_API_KEY = st.secrets["GEMINI_API_KEY"]
 client = genai.Client(api_key=GEMINI_API_KEY)
@@ -22,7 +19,6 @@ st.write("Upload your speech and get AI-powered feedback")
 
 def analyze_knowledge_relevancy(audio_data, title):
     prompt = f"""As an expert in content analysis, evaluate this speech titled '{title}' focusing ONLY on:
-
     1. Knowledge Depth:
     - Topic expertise level
     - Accuracy of information
@@ -32,7 +28,6 @@ def analyze_knowledge_relevancy(audio_data, title):
     - Alignment with topic
     - Appropriate examples
     - Target audience fit
-
     Provide a structured analysis with specific examples from the speech."""
 
     contents = [
@@ -46,10 +41,10 @@ def analyze_knowledge_relevancy(audio_data, title):
     ]
 
     response = client.models.generate_content(
-        model="gemini-2.0-flash",
+        model="gemini-2.0-pro-exp-02-05",
         contents=contents,
         config=types.GenerateContentConfig(
-            temperature=0.7,
+            temperature=0.4,
             top_p=0.95,
             top_k=40,
             max_output_tokens=8192,
@@ -60,7 +55,6 @@ def analyze_knowledge_relevancy(audio_data, title):
 
 def analyze_emotional_delivery(audio_data, title):
     prompt = f"""As an expert in public speaking delivery, analyze this speech titled '{title}' focusing ONLY on:
-
     1. Emotional Expression:
     - Voice modulation
     - Emotional engagement
@@ -71,7 +65,6 @@ def analyze_emotional_delivery(audio_data, title):
     - Use of pauses
     - Filler words
     - Voice clarity
-
     Provide specific examples and timestamps where possible."""
 
     contents = [
@@ -85,10 +78,10 @@ def analyze_emotional_delivery(audio_data, title):
     ]
 
     response = client.models.generate_content(
-        model="gemini-2.0-flash",
+        model="gemini-2.0-pro-exp-02-05",
         contents=contents,
         config=types.GenerateContentConfig(
-            temperature=0.7,
+            temperature=0.4,
             top_p=0.95,
             top_k=40,
             max_output_tokens=8192,
@@ -99,13 +92,10 @@ def analyze_emotional_delivery(audio_data, title):
 
 def generate_final_analysis(knowledge_analysis, emotional_analysis):
     prompt_final = f"""As a comprehensive public speaking coach, analyze these two detailed evaluations:
-
     Knowledge Analysis:
     {knowledge_analysis}
-
     Emotional Delivery Analysis:
     {emotional_analysis}
-
     Provide:
     1. Overall Score (0-100)
     2. Key Strengths (Top 3)
@@ -127,7 +117,7 @@ def generate_final_analysis(knowledge_analysis, emotional_analysis):
     ]
 
     response = client.models.generate_content(
-        model="gemini-2.0-flash",
+        model="gemini-2.0-pro-exp-02-05",
         contents=contents,
         config=types.GenerateContentConfig(
             temperature=0.7,
@@ -138,7 +128,7 @@ def generate_final_analysis(knowledge_analysis, emotional_analysis):
     )
     
     return response.text
-
+    
 def parallel_analysis(audio_data, title):
     with concurrent.futures.ThreadPoolExecutor() as executor:
         # Submit both analysis tasks
@@ -170,118 +160,52 @@ def parallel_analysis(audio_data, title):
 
 # Main interface
 title = st.text_input("Speech Title/Topic:", placeholder="e.g., Introduction to Machine Learning")
-# Input method selection
-input_method = st.radio("Choose input method:", ["Upload Audio File", "Record Speech"], key="input_method_radio")
-
-audio_data = None
-audio_path = None
 
-if input_method == "Upload Audio File":
-    uploaded_file = st.file_uploader("Upload your speech (WAV, MP3, M4A)", type=["wav", "mp3", "m4a"], key="speech_file_uploader")
-    if uploaded_file:
-        st.audio(uploaded_file, key="uploaded_audio_player")
-        if title and uploaded_file:
-            if st.button("Analyze Speech", key="analyze_uploaded_speech_button"):
-                with st.spinner("Processing your speech..."):
-                    # Save and process audio
-                    with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp_file:
-                        audio_data = uploaded_file.read()
-                        tmp_file.write(audio_data)
-                        audio_path = tmp_file.name
-                    
-                    try:
-                        # Run parallel analysis
-                        knowledge, emotional, final = parallel_analysis(audio_data, title)
-                        
-                        # Download options
-                        col1, col2, col3 = st.columns(3)
-                        with col1:
-                            st.download_button(
-                                "Download Knowledge Analysis",
-                                knowledge,
-                                file_name=f"knowledge_analysis_{title}.txt",
-                                key="download_knowledge_button"
-                            )
-                        with col2:
-                            st.download_button(
-                                "Download Emotional Analysis",
-                                emotional,
-                                file_name=f"emotional_analysis_{title}.txt",
-                                key="download_emotional_button"
-                            )
-                        with col3:
-                            st.download_button(
-                                "Download Final Analysis",
-                                final,
-                                file_name=f"final_analysis_{title}.txt",
-                                key="download_final_button"
-                            )
-                    
-                    except Exception as e:
-                        st.error(f"Error during analysis: {str(e)}")
-                        if "API key" in str(e):
-                            st.warning("Please check your Google API key configuration.")
-                    
-                    finally:
-                        if os.path.exists(audio_path):
-                            os.unlink(audio_path)
-        else:
-            st.info("Please provide both a title and upload your speech recording to begin.")
-
-else:  # Record Speech
-    st.write("Record your speech directly:")
-    audio_bytes = st_audiorecorder(pause_threshold=2.0, sample_rate=44100, key="speech_recorder")
-    
-    if audio_bytes and title:
-        # Convert audio bytes to WAV format using pydub
-        audio_segment = pydub.AudioSegment.from_wav(audio_bytes)
-        
-        st.audio(audio_bytes, key="recorded_audio_player")
-        
-        if st.button("Analyze Recorded Speech", key="analyze_recorded_speech_button"):
-            with st.spinner("Processing your speech..."):
-                # Save recorded audio to temporary file
-                with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp_file:
-                    audio_segment.export(tmp_file.name, format="wav")
-                    audio_path = tmp_file.name
-                
-                try:
-                    # Run parallel analysis
-                    knowledge, emotional, final = parallel_analysis(audio_bytes, title)
-                    
-                    # Download options
-                    col1, col2, col3 = st.columns(3)
-                    with col1:
-                        st.download_button(
-                            "Download Knowledge Analysis",
-                            knowledge,
-                            file_name=f"knowledge_analysis_{title}.txt",
-                            key="download_recorded_knowledge_button"
-                        )
-                    with col2:
-                        st.download_button(
-                            "Download Emotional Analysis",
-                            emotional,
-                            file_name=f"emotional_analysis_{title}.txt",
-                            key="download_recorded_emotional_button"
-                        )
-                    with col3:
-                        st.download_button(
-                            "Download Final Analysis",
-                            final,
-                            file_name=f"final_analysis_{title}.txt",
-                            key="download_recorded_final_button"
-                        )
-                
-                except Exception as e:
-                    st.error(f"Error during analysis: {str(e)}")
-                    if "API key" in str(e):
-                        st.warning("Please check your Google API key configuration.")
+uploaded_file = st.file_uploader("Upload your speech (WAV, MP3, M4A)", type=["wav", "mp3", "m4a"])
+if uploaded_file:
+    st.audio(uploaded_file)
+
+if title and uploaded_file:
+    if st.button("Analyze Speech"):
+        with st.spinner("Processing your speech..."):
+            # Save and process audio
+            with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp_file:
+                audio_data = uploaded_file.read()
+                tmp_file.write(audio_data)
+                audio_path = tmp_file.name
+            
+            try:
+                # Run parallel analysis
+                knowledge, emotional, final = parallel_analysis(audio_data, title)
                 
-                finally:
-                    if os.path.exists(audio_path):
-                        os.unlink(audio_path)
-    elif not title and audio_bytes:
-        st.info("Please provide a title for your speech to begin analysis.")
-    elif not audio_bytes:
-        st.info("Record your speech using the microphone button above.")
+                # Download options
+                col1, col2, col3 = st.columns(3)
+                with col1:
+                    st.download_button(
+                        "Download Knowledge Analysis",
+                        knowledge,
+                        file_name=f"knowledge_analysis_{title}.txt"
+                    )
+                with col2:
+                    st.download_button(
+                        "Download Emotional Analysis",
+                        emotional,
+                        file_name=f"emotional_analysis_{title}.txt"
+                    )
+                with col3:
+                    st.download_button(
+                        "Download Final Analysis",
+                        final,
+                        file_name=f"final_analysis_{title}.txt"
+                    )
+            
+            except Exception as e:
+                st.error(f"Error during analysis: {str(e)}")
+                if "API key" in str(e):
+                    st.warning("Please check your Google API key configuration.")
+            
+            finally:
+                if os.path.exists(audio_path):
+                    os.unlink(audio_path)
+else:
+    st.info("Please provide both a title and upload your speech recording to begin.")