Update app.py

app.py

@@ -1,157 +1,130 @@
 import streamlit as st
-import whisper
 import tempfile
 import os
-import time
 import re
+import time
 import torch
 from pydub import AudioSegment
+from transformers import pipeline
 from openpyxl import Workbook
-from openpyxl.styles import Font
 from docx import Document
-from docx.shared import Pt
-from docx.enum.text import WD_ALIGN_PARAGRAPH
 from io import BytesIO
 
-# ---------------------------------------------------
-# PAGE CONFIG
-# ---------------------------------------------------
-st.set_page_config(
-    page_title="RecToText Pro",
-    layout="wide",
-    page_icon="🎤"
-)
+st.set_page_config(page_title="RecToText Pro", layout="wide")
 
 st.title("🎤 RecToText Pro")
-st.caption("Stable Production Version | CPU Optimized")
-
-# ---------------------------------------------------
-# SIDEBAR
-# ---------------------------------------------------
-model_option = st.sidebar.selectbox(
-    "Select Whisper Model",
-    ["base"]  # Force base for stability
-)
+st.caption("Stable HuggingFace Build Version")
 
-output_mode = st.sidebar.radio(
-    "Output Format",
-    ["Roman Urdu", "English"]
-)
-
-# ---------------------------------------------------
-# LOAD MODEL (FORCE CPU)
-# ---------------------------------------------------
+# -------------------------
+# LOAD MODEL (HF PIPELINE)
+# -------------------------
 @st.cache_resource
-def load_model():
-    return whisper.load_model("base", device="cpu")
-
-# ---------------------------------------------------
-# CLEAN TEXT
-# ---------------------------------------------------
+def load_asr():
+    return pipeline(
+        "automatic-speech-recognition",
+        model="openai/whisper-base",
+        device=-1  # CPU
+    )
+
+asr = load_asr()
+
+# -------------------------
+# TEXT CLEANING
+# -------------------------
 def clean_text(text):
-    filler_words = ["um", "hmm", "acha", "matlab", "uh"]
-    pattern = r'\b(?:' + '|'.join(filler_words) + r')\b'
-    text = re.sub(pattern, '', text, flags=re.IGNORECASE)
-    text = re.sub(r'\s+', ' ', text).strip()
-    return text
-
-# ---------------------------------------------------
-# ROMAN URDU
-# ---------------------------------------------------
-def convert_to_roman_urdu(text):
+    filler = ["um", "hmm", "acha", "matlab"]
+    pattern = r'\b(?:' + '|'.join(filler) + r')\b'
+    text = re.sub(pattern, "", text, flags=re.IGNORECASE)
+    return re.sub(r'\s+', ' ', text).strip()
+
+# -------------------------
+# ROMAN URDU BASIC
+# -------------------------
+def convert_to_roman(text):
     replacements = {
         "ہے": "hai",
         "میں": "main",
         "اور": "aur",
         "کیا": "kya"
     }
-    for urdu, roman in replacements.items():
-        text = text.replace(urdu, roman)
+    for k, v in replacements.items():
+        text = text.replace(k, v)
     return text
 
-# ---------------------------------------------------
+# -------------------------
 # EXCEL EXPORT
-# ---------------------------------------------------
-def create_excel(text):
+# -------------------------
+def export_excel(text):
     wb = Workbook()
     ws = wb.active
     ws.append(["Transcription"])
-    ws["A1"].font = Font(bold=True)
     ws.append([text])
-
     buffer = BytesIO()
     wb.save(buffer)
     buffer.seek(0)
     return buffer
 
-# ---------------------------------------------------
+# -------------------------
 # WORD EXPORT
-# ---------------------------------------------------
-def create_word(text):
+# -------------------------
+def export_word(text):
     doc = Document()
     doc.add_heading("Lecture Transcription", level=1)
     doc.add_paragraph(text)
-
     buffer = BytesIO()
     doc.save(buffer)
     buffer.seek(0)
     return buffer
 
-# ---------------------------------------------------
+# -------------------------
 # FILE UPLOADER
-# ---------------------------------------------------
-uploaded_file = st.file_uploader(
-    "Upload Lecture (.mp3, .wav, .m4a, .aac)",
+# -------------------------
+uploaded = st.file_uploader(
+    "Upload Audio (.mp3, .wav, .m4a, .aac)",
     type=["mp3", "wav", "m4a", "aac"]
 )
 
-if uploaded_file:
+output_mode = st.radio("Output Format", ["English", "Roman Urdu"])
 
+if uploaded:
     try:
-        st.audio(uploaded_file)
+        st.audio(uploaded)
 
         with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp:
-            ext = uploaded_file.name.split(".")[-1]
-            audio = AudioSegment.from_file(uploaded_file, format=ext)
+            ext = uploaded.name.split(".")[-1]
+            audio = AudioSegment.from_file(uploaded, format=ext)
             audio.export(tmp.name, format="wav")
             temp_path = tmp.name
 
-        model = load_model()
+        start = time.time()
 
-        with st.spinner("Transcribing safely on CPU..."):
-            result = model.transcribe(temp_path)
+        with st.spinner("Transcribing..."):
+            result = asr(temp_path)
 
         os.remove(temp_path)
 
         text = result["text"]
-        cleaned = clean_text(text)
+        text = clean_text(text)
 
         if output_mode == "Roman Urdu":
-            cleaned = convert_to_roman_urdu(cleaned)
+            text = convert_to_roman(text)
 
         st.success("Transcription Completed ✅")
+        st.text_area("Output", text, height=300)
 
-        st.text_area("Output", cleaned, height=300)
-
-        excel_file = create_excel(cleaned)
-        word_file = create_word(cleaned)
+        excel_file = export_excel(text)
+        word_file = export_word(text)
 
         col1, col2 = st.columns(2)
 
         with col1:
-            st.download_button(
-                "Download Excel",
-                excel_file,
-                "RecToText.xlsx"
-            )
+            st.download_button("Download Excel", excel_file, "RecToText.xlsx")
 
         with col2:
-            st.download_button(
-                "Download Word",
-                word_file,
-                "RecToText.docx"
-            )
+            st.download_button("Download Word", word_file, "RecToText.docx")
+
+        st.write(f"Processing Time: {round(time.time()-start,2)} sec")
 
     except Exception as e:
-        st.error("Processing Error Occurred.")
+        st.error("Error Occurred")
         st.exception(e)