Update app.py

app.py

@@ -1,144 +1,54 @@
 import gradio as gr
 import cv2
 import numpy as np
-import tempfile
+import json
 import os
-from PIL import Image
+import tempfile
 
-def extract_subtitles(video_file, language, start_time, end_time, box_x, box_y, box_w, box_h):
-    """
-    Video ထဲက hardcoded subtitles တွေကို ထုတ်ပေးမယ့် function
-    """
-    # Video ဖိုင်ကိုဖွင့်မယ်
-    cap = cv2.VideoCapture(video_file)
-    
-    # Frame rate သတ်မှတ်မယ်
-    fps = cap.get(cv2.CAP_PROP_FPS)
-    
-    # Start/End time ကို frame number အဖြစ်ပြောင်းမယ်
-    start_frame = int(start_time * fps)
-    end_frame = int(end_time * fps)
-    
-    # စာတန်းတွေသိမ်းမယ့် list
-    subtitles = []
-    last_text = ""
-    current_start = start_time
-    
-    # Frame အလိုက်ဖတ်မယ် (1 FPS)
-    cap.set(cv2.CAP_PROP_POS_FRAMES, start_frame)
-    
-    for frame_num in range(start_frame, end_frame, int(fps)):
-        ret, frame = cap.read()
-        if not ret:
-            break
-            
-        # ရွေးထားတဲ့နေရာကိုဖြတ်မယ်
-        h, w = frame.shape[:2]
-        x1 = int((box_x / 100) * w)
-        y1 = int((box_y / 100) * h)
-        x2 = int(((box_x + box_w) / 100) * w)
-        y2 = int(((box_y + box_h) / 100) * h)
-        
-        crop = frame[y1:y2, x1:x2]
-        
-        # ဒီနေရာမှာ OCR လုပ်ဖို့လိုမယ် (Tesseract/EasyOCR/PaddleOCR)
-        # ယခု ဥပမာအနေနဲ့ စာသားအတုပဲထုတ်ထားတယ်
-        text = f"Sample OCR text at {frame_num/fps:.1f}s"
-        
-        # စာသားတူရင် merge လုပ်မယ်
-        if text and text != last_text:
-            if last_text:
-                subtitles.append({
-                    'start': current_start,
-                    'end': frame_num / fps,
-                    'text': last_text
-                })
-            current_start = frame_num / fps
-            last_text = text
-    
-    # နောက်ဆုံးစာတန်းကိုထည့်မယ်
-    if last_text:
-        subtitles.append({
-            'start': current_start,
-            'end': end_time,
-            'text': last_text
-        })
-    
-    cap.release()
-    
-    # SRT format ပြောင်းမယ်
+def test_function(video):
+    return "Video uploaded: " + str(video)
+
+def extract_subtitles(video, language, start, end, x, y, w, h):
+    # Create sample subtitles
+    subtitles = [
+        {"start": 0, "end": 2, "text": "Hello"},
+        {"start": 2, "end": 4, "text": "Welcome to video OCR"},
+        {"start": 4, "end": 6, "text": "This is a test"}
+    ]
+    
+    # Generate SRT
     srt_content = ""
     for i, sub in enumerate(subtitles, 1):
-        start = format_time(sub['start'])
-        end = format_time(sub['end'])
-        srt_content += f"{i}\n{start} --> {end}\n{sub['text']}\n\n"
+        srt_content += f"{i}\n00:00:{sub['start']:02d},000 --> 00:00:{sub['end']:02d},000\n{sub['text']}\n\n"
     
-    # SRT ဖိုင်သိမ်းမယ်
-    srt_path = "extracted_subtitles.srt"
-    with open(srt_path, "w", encoding="utf-8") as f:
+    # Save files
+    srt_path = "subtitles.srt"
+    with open(srt_path, "w") as f:
         f.write(srt_content)
     
-    # စာတန်းစာရင်းကို Dataframe အတွက်ပြင်ဆင်မယ်
-    subtitle_list = [[format_time(s['start']), format_time(s['end']), s['text']] for s in subtitles]
+    # Display list
+    display_list = [[f"00:00:{s['start']:02d}", f"00:00:{s['end']:02d}", s['text']] for s in subtitles]
     
-    return srt_path, subtitle_list
-
-def format_time(seconds):
-    """အချိန်ကို SRT format (HH:MM:SS,mmm) ပြောင်းမယ်"""
-    h = int(seconds // 3600)
-    m = int((seconds % 3600) // 60)
-    s = int(seconds % 60)
-    ms = int((seconds - int(seconds)) * 1000)
-    return f"{h:02d}:{m:02d}:{s:02d},{ms:03d}"
+    return srt_path, display_list
 
-# Gradio Interface ဆောက်မယ်
-with gr.Blocks(theme=gr.themes.Soft(), title="Video Hardsub OCR") as demo:
-    gr.Markdown("# 🎬 Video Hardsub OCR Extractor")
-    gr.Markdown("Video ထဲက hardcoded စာတန်းတွေကို ထုတ်ယူပါ။")
+# Simple interface
+with gr.Blocks(title="Video OCR") as demo:
+    gr.Markdown("# Video OCR Extractor")
     
     with gr.Row():
-        with gr.Column(scale=1):
-            # Video Input
+        with gr.Column():
             video_input = gr.Video(label="Upload Video")
-            
-            # Language Selection
-            language = gr.Dropdown(
-                choices=["English", "Thai", "Japanese", "Chinese (Simplified)", "Chinese (Traditional)"],
-                label="Language",
-                value="English"
-            )
-            
-            # Time Range
-            with gr.Row():
-                start_time = gr.Number(label="Start Time (s)", value=0)
-                end_time = gr.Number(label="End Time (s)", value=10)
-            
-            # Subtitle Area (Box)
-            with gr.Accordion("Subtitle Area Adjustment", open=True):
-                gr.Markdown("စာတန်းရှိတဲ့နေရာကို ချိန်ညှိပါ")
-                box_x = gr.Slider(0, 100, value=10, label="X Position (%)")
-                box_y = gr.Slider(0, 100, value=80, label="Y Position (%)")
-                box_w = gr.Slider(0, 100, value=80, label="Width (%)")
-                box_h = gr.Slider(0, 100, value=15, label="Height (%)")
-            
-            process_btn = gr.Button("Process Video", variant="primary")
+            language = gr.Dropdown(["English", "Thai", "Japanese", "Chinese"], label="Language")
+            process_btn = gr.Button("Process", variant="primary")
         
-        with gr.Column(scale=1):
-            # Output
-            video_output = gr.Video(label="Video Preview")
-            subtitle_output = gr.Dataframe(
-                headers=["Start", "End", "Text"],
-                label="Extracted Subtitles",
-                wrap=True
-            )
+        with gr.Column():
+            subtitle_output = gr.Dataframe(headers=["Start", "End", "Text"], label="Subtitles")
             file_output = gr.File(label="Download SRT")
     
-    # Process function
     process_btn.click(
         fn=extract_subtitles,
-        inputs=[video_input, language, start_time, end_time, box_x, box_y, box_w, box_h],
+        inputs=[video_input, language, gr.State(0), gr.State(10), gr.State(10), gr.State(80), gr.State(80), gr.State(15)],
         outputs=[file_output, subtitle_output]
     )
 
-# Launch
 demo.launch()