Update app.py

app.py

@@ -353,8 +353,14 @@ def create_subtitle_clip_pil(text, start_time, end_time, video_width, video_heig
 
 def process_entry(entry, i, video_width, video_height, add_voiceover, target_language, font_path, speaker_sample_paths=None):
     logger.debug(f"Processing entry {i}: {entry}")
+    error_message = None
 
-    txt_clip = create_subtitle_clip_pil(entry["translated"], entry["start"], entry["end"], video_width, video_height, font_path)
+    try:
+        txt_clip = create_subtitle_clip_pil(entry["translated"], entry["start"], entry["end"], video_width, video_height, font_path)
+    except Exception as e:
+        error_message = f"❌ Failed to create subtitle clip for entry {i}: {e}"
+        logger.error(error_message)
+        txt_clip = None
 
     audio_segment = None
     if add_voiceover:
@@ -362,8 +368,7 @@ def process_entry(entry, i, video_width, video_height, add_voiceover, target_lan
             segment_audio_path = f"segment_{i}_voiceover.wav"
             desired_duration = entry["end"] - entry["start"]
             speaker = entry.get("speaker", "default")
-            speaker_wav_path = f"speaker_{speaker}_sample.wav" 
-            #speaker_sample_paths.get(speaker_id, "speaker_default.wav") if speaker_sample_paths else "speaker_default.wav"
+            speaker_wav_path = f"speaker_{speaker}_sample.wav"
 
             generate_voiceover_clone([entry], desired_duration, target_language, speaker_wav_path, segment_audio_path)
 
@@ -374,66 +379,61 @@ def process_entry(entry, i, video_width, video_height, add_voiceover, target_lan
             logger.debug(f"Audio clip duration: {audio_clip.duration}, Desired duration: {desired_duration}")
 
             if audio_clip.duration < desired_duration:
-                # Pad with silence if audio is too short
                 silence_duration = desired_duration - audio_clip.duration
-    
-                # Concatenate the original audio and silence
                 audio_clip = concatenate_audioclips([audio_clip, silence(duration=silence_duration)])
                 logger.info(f"Padded audio with {silence_duration} seconds of silence.")
-            
-                audio_segment = audio_clip.set_start(entry["start"]).set_duration(desired_duration)
+
+            audio_segment = audio_clip.set_start(entry["start"]).set_duration(desired_duration)
+
         except Exception as e:
-            logger.error(f"\u274c Failed to generate audio segment for entry {i}: {e}")
+            err = f"❌ Failed to generate audio segment for entry {i}: {e}"
+            logger.error(err)
+            error_message = error_message + " | " + err if error_message else err
             audio_segment = None
 
-    return i, txt_clip, audio_segment
+    return i, txt_clip, audio_segment, error_message
 
-def add_transcript_voiceover(video_path, translated_json, output_path, add_voiceover=False, target_language="en", speaker_sample_paths=None):
-    """
-    Add transcript and voiceover to a video, segment by segment.
-    """
-    video = VideoFileClip(video_path)
-    font_path = "./NotoSansSC-Regular.ttf"
+def process_entry(entry, i, video_width, video_height, add_voiceover, target_language, font_path, speaker_sample_paths=None):
+    logger.debug(f"Processing entry {i}: {entry}")
+    error_message = None
 
-    text_clips = []
-    audio_segments = []
+    try:
+        txt_clip = create_subtitle_clip_pil(entry["translated"], entry["start"], entry["end"], video_width, video_height, font_path)
+    except Exception as e:
+        error_message = f"❌ Failed to create subtitle clip for entry {i}: {e}"
+        logger.error(error_message)
+        txt_clip = None
 
-    with concurrent.futures.ThreadPoolExecutor() as executor:
-        futures = [executor.submit(process_entry, entry, i, video.w, video.h, add_voiceover, target_language, font_path, speaker_sample_paths)
-                   for i, entry in enumerate(translated_json)]
-
-        # Collect results with original index i
-        results = []
-        for future in concurrent.futures.as_completed(futures):
-            try:
-                i, txt_clip, audio_segment = future.result()
-                results.append((i, txt_clip, audio_segment))
-            except Exception as e:
-                logger.error(f"Error processing entry: {e}")
-                
-    # Sort by original index i
-    results.sort(key=lambda x: x[0])
-    
-    # Extract sorted clips
-    text_clips = [clip for i, clip, segment in results]
+    audio_segment = None
+    if add_voiceover:
+        try:
+            segment_audio_path = f"segment_{i}_voiceover.wav"
+            desired_duration = entry["end"] - entry["start"]
+            speaker = entry.get("speaker", "default")
+            speaker_wav_path = f"speaker_{speaker}_sample.wav"
 
-    final_video = CompositeVideoClip([video] + text_clips)
+            generate_voiceover_clone([entry], desired_duration, target_language, speaker_wav_path, segment_audio_path)
 
-    logger.info("Methods in CompositeVideoClip:")
-    for method in dir(final_video):
-        logger.info(method)
-    
-    if add_voiceover:
-        audio_segments = [segment for i, clip, segment in results if segment is not None]
-        final_audio = CompositeAudioClip(audio_segments)  # Critical fix
-        final_audio = final_audio.set_duration(video.duration)
+            if not os.path.exists(segment_audio_path):
+                raise FileNotFoundError(f"Voiceover file not generated at: {segment_audio_path}")
+
+            audio_clip = AudioFileClip(segment_audio_path)
+            logger.debug(f"Audio clip duration: {audio_clip.duration}, Desired duration: {desired_duration}")
 
-        final_video = final_video.set_audio(final_audio)
+            if audio_clip.duration < desired_duration:
+                silence_duration = desired_duration - audio_clip.duration
+                audio_clip = concatenate_audioclips([audio_clip, silence(duration=silence_duration)])
+                logger.info(f"Padded audio with {silence_duration} seconds of silence.")
+
+            audio_segment = audio_clip.set_start(entry["start"]).set_duration(desired_duration)
 
-    logger.info(f"Saving the final video to: {output_path}")
-    final_video.write_videofile(output_path, codec="libx264", audio_codec="aac")
+        except Exception as e:
+            err = f"❌ Failed to generate audio segment for entry {i}: {e}"
+            logger.error(err)
+            error_message = error_message + " | " + err if error_message else err
+            audio_segment = None
 
-    logger.info("Video processing completed successfully.")
+    return i, txt_clip, audio_segment, error_message
 
 import os
 import traceback