Create app.py

app.py

@@ -0,0 +1,76 @@
+import gradio as gr
+import os
+from videocr import save_subtitles_to_file
+
+# Define OCR function
+def run_video_ocr(input_video, output_file_name, language_code, use_gpu, start_time, end_time, confidence_threshold, similarity_threshold, frames_to_skip, crop_x, crop_y, crop_width, crop_height):
+    try:
+        # Ensure the output directory exists
+        persistent_dir = '/persistent'
+        if not os.path.exists(persistent_dir):
+            os.makedirs(persistent_dir)
+        
+        # Define full path for the output file
+        output_path = os.path.join(persistent_dir, output_file_name)
+
+        # Save the subtitles to file
+        save_subtitles_to_file(
+            input_video.name,  # The uploaded video file
+            output_path,       # Output .srt file path
+            lang=language_code,
+            use_gpu=use_gpu,
+            time_start=start_time,
+            time_end=end_time,
+            conf_threshold=confidence_threshold,
+            sim_threshold=similarity_threshold,
+            frames_to_skip=frames_to_skip,
+            crop_x=crop_x,
+            crop_y=crop_y,
+            crop_width=crop_width,
+            crop_height=crop_height
+        )
+
+        return f"Subtitle extraction completed! File saved to {output_path}"
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+# Define Gradio interface
+def video_ocr_interface():
+    with gr.Blocks() as demo:
+        with gr.Row():
+            input_video = gr.File(label="Upload Video", type="file")
+            output_file_name = gr.Textbox(label="Output File Name (.srt)", value="subtitle.srt")
+            language_code = gr.Textbox(label="Language Code", value="ch")
+            use_gpu = gr.Checkbox(label="Use GPU", value=True)
+        
+        with gr.Row():
+            start_time = gr.Textbox(label="Start Time (HH:MM:SS)", value="00:00:00")
+            end_time = gr.Textbox(label="End Time (HH:MM:SS)", value="")
+        
+        with gr.Row():
+            confidence_threshold = gr.Slider(label="Confidence Threshold", minimum=0, maximum=100, value=75)
+            similarity_threshold = gr.Slider(label="Similarity Threshold", minimum=0, maximum=100, value=80)
+        
+        with gr.Row():
+            frames_to_skip = gr.Slider(label="Frames to Skip", minimum=0, maximum=10, value=0)
+            crop_x = gr.Number(label="Crop X", value=0)
+            crop_y = gr.Number(label="Crop Y", value=0)
+            crop_width = gr.Number(label="Crop Width", value=0)
+            crop_height = gr.Number(label="Crop Height", value=0)
+
+        submit_btn = gr.Button("Start OCR")
+
+        output_label = gr.Textbox(label="Status", interactive=False)
+
+        # Define what happens when the button is clicked
+        submit_btn.click(
+            fn=run_video_ocr,
+            inputs=[input_video, output_file_name, language_code, use_gpu, start_time, end_time, confidence_threshold, similarity_threshold, frames_to_skip, crop_x, crop_y, crop_width, crop_height],
+            outputs=output_label
+        )
+    
+    return demo
+
+# Launch the Gradio interface
+demo = video_ocr_interface()
+demo.launch()