Update app.py

app.py

@@ -9,43 +9,44 @@ from openpyxl import Workbook
 from openpyxl.styles import Font
 from io import BytesIO
 
-# ---------------------------
+# ---------------------------------------------------
 # PAGE CONFIG
-# ---------------------------
+# ---------------------------------------------------
 st.set_page_config(
     page_title="RecToText Pro",
     layout="wide",
     page_icon="🎤"
 )
 
-# ---------------------------
-# SIDEBAR
-# ---------------------------
+# ---------------------------------------------------
+# SIDEBAR SETTINGS
+# ---------------------------------------------------
 st.sidebar.title("⚙️ Settings")
+
 model_option = st.sidebar.selectbox(
     "Select Whisper Model",
     ["base", "small"]
 )
 
 output_mode = st.sidebar.radio(
-    "Output Format",
+    "Output Language Output",
     ["Roman Urdu", "English"]
 )
 
 if st.sidebar.button("🧹 Clear Session"):
     st.session_state.clear()
-    st.experimental_rerun()
+    st.rerun()
 
-# ---------------------------
+# ---------------------------------------------------
 # HEADER
-# ---------------------------
+# ---------------------------------------------------
 st.markdown("<h1 style='text-align:center;'>🎤 RecToText Pro</h1>", unsafe_allow_html=True)
-st.markdown("<p style='text-align:center;'>Intelligent Urdu + English Lecture Transcriber</p>", unsafe_allow_html=True)
+st.markdown("<p style='text-align:center;'>AI-Powered Urdu + English Lecture Transcriber</p>", unsafe_allow_html=True)
 st.divider()
 
-# ---------------------------
+# ---------------------------------------------------
 # FUNCTIONS
-# ---------------------------
+# ---------------------------------------------------
 
 @st.cache_resource
 def load_model(model_size):
@@ -59,22 +60,19 @@ def clean_text(text):
     return text
 
 def convert_to_roman_urdu(text):
-    # Basic placeholder conversion logic
     replacements = {
         "ہے": "hai",
         "میں": "main",
         "اور": "aur",
         "کیا": "kya",
-        "آپ": "aap"
+        "آپ": "aap",
+        "کی": "ki",
+        "کا": "ka"
     }
     for urdu, roman in replacements.items():
         text = text.replace(urdu, roman)
     return text
 
-def process_audio(file_path, model):
-    result = model.transcribe(file_path)
-    return result
-
 def create_excel(segments):
     wb = Workbook()
     ws = wb.active
@@ -97,12 +95,13 @@ def create_excel(segments):
     excel_buffer.seek(0)
     return excel_buffer
 
-# ---------------------------
-# FILE UPLOADER
-# ---------------------------
+# ---------------------------------------------------
+# FILE UPLOADER (AAC SUPPORTED)
+# ---------------------------------------------------
 uploaded_file = st.file_uploader(
-    "Upload Lecture Recording (.mp3, .wav, .m4a)",
-    type=["mp3", "wav", "m4a"]
+    "Upload Lecture Recording (.mp3, .wav, .m4a, .aac)",
+    type=["mp3", "wav", "m4a", "aac"],
+    help="Supported formats: mp3, wav, m4a, aac"
 )
 
 if uploaded_file:
@@ -110,18 +109,19 @@ if uploaded_file:
     st.audio(uploaded_file)
 
     with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp:
-        audio = AudioSegment.from_file(uploaded_file)
+        file_extension = uploaded_file.name.split(".")[-1]
+        audio = AudioSegment.from_file(uploaded_file, format=file_extension)
         audio.export(tmp.name, format="wav")
         temp_audio_path = tmp.name
 
-    st.info("Loading model...")
+    st.info("Loading Whisper model...")
     model = load_model(model_option)
 
     progress = st.progress(0)
     start_time = time.time()
 
-    with st.spinner("Transcribing..."):
-        result = process_audio(temp_audio_path, model)
+    with st.spinner("Transcribing... Please wait."):
+        result = model.transcribe(temp_audio_path)
         progress.progress(100)
 
     end_time = time.time()
@@ -130,54 +130,52 @@ if uploaded_file:
 
     detected_lang = result.get("language", "Unknown")
     segments = result["segments"]
-
     full_text = result["text"]
+
     cleaned_text = clean_text(full_text)
 
     if output_mode == "Roman Urdu":
         cleaned_text = convert_to_roman_urdu(cleaned_text)
-    else:
-        cleaned_text = cleaned_text
 
     word_count = len(cleaned_text.split())
     processing_time = round(end_time - start_time, 2)
 
-    # ---------------------------
+    # ---------------------------------------------------
     # DISPLAY RESULTS
-    # ---------------------------
+    # ---------------------------------------------------
     col1, col2 = st.columns(2)
 
     with col1:
         st.subheader("📜 Raw Transcription")
-        st.text_area("", full_text, height=300)
+        st.text_area("", full_text, height=350)
 
     with col2:
         st.subheader("✨ Cleaned Output")
-        st.text_area("", cleaned_text, height=300)
+        st.text_area("", cleaned_text, height=350)
 
     st.divider()
 
     st.write(f"**Detected Language:** {detected_lang}")
     st.write(f"**Word Count:** {word_count}")
-    st.write(f"**Processing Time:** {processing_time} sec")
+    st.write(f"**Processing Time:** {processing_time} seconds")
 
-    # ---------------------------
+    # ---------------------------------------------------
     # EXCEL DOWNLOAD
-    # ---------------------------
+    # ---------------------------------------------------
     excel_file = create_excel(segments)
 
     st.download_button(
         label="📥 Download Excel File",
         data=excel_file,
-        file_name="transcription.xlsx",
+        file_name="RecToText_Transcription.xlsx",
         mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
     )
 
-# ---------------------------
+# ---------------------------------------------------
 # FOOTER
-# ---------------------------
+# ---------------------------------------------------
 st.divider()
 st.markdown(
-    "<p style='text-align:center; font-size:12px;'>Developed with ❤️ using Whisper & Streamlit</p>",
+    "<p style='text-align:center; font-size:12px;'>Developed using Whisper & Streamlit | RecToText Pro</p>",
     unsafe_allow_html=True
 )