Update app.py

app.py

@@ -199,8 +199,29 @@ def update_translations(file, edited_table):
     except Exception as e:
         raise ValueError(f"Error updating translations: {e}")
 
+def generate_voiceover(translated_json, language, output_audio_path):
+    from gtts import gTTS
+    
+    # Concatenate translated text into a single string
+    full_text = " ".join(entry["translated"] for entry in translated_json)
+    
+    # Generate speech
+    tts = gTTS(text=full_text, lang=language)
+    tts.save(output_audio_path)
+
+def replace_audio_in_video(video_path, new_audio_path, final_video_path):
+    import moviepy.editor as mp
+    
+    video = mp.VideoFileClip(video_path)
+    new_audio = mp.AudioFileClip(new_audio_path)
+    
+    # Set the new audio
+    video = video.set_audio(new_audio)
+    
+    # Save the final output
+    video.write_videofile(final_video_path, codec="libx264", audio_codec="aac")
 
-def upload_and_manage(file, language):
+def upload_and_manage(file, language, mode="transcription"):
     if file is None:
         return None, [], None, "No file uploaded. Please upload a video/audio file."
 
@@ -210,19 +231,25 @@ def upload_and_manage(file, language):
         # Define paths for audio and output files
         audio_path = "audio.wav"
         output_video_path = "output_video.mp4"
+        voiceover_path = "voiceover.wav"
 
         list_available_fonts()
 
-        # Transcribe audio from uploaded media file and get timestamps
-        transcrption_json = transcribe_video(file.name)
+        # Step 1: Transcribe audio from uploaded media file and get timestamps
+        transcription_json = transcribe_video(file.name)
 
-        # Translate the transcription
-        translated_json = translate_text(transcrption_json, language)
-        
-        # Add transcript to video based on timestamps
+        # Step 2: Translate the transcription
+        translated_json = translate_text(transcription_json, language)
+
+        # Step 3: Add transcript to video based on timestamps
         add_transcript_to_video(file.name, translated_json, output_video_path)
 
-        # Convert the translated JSON into a format for the editable table
+        # Step 4 (Optional): Generate voiceover if mode is "transcription_voiceover"
+        if mode == "transcription_voiceover":
+            generate_voiceover(translated_json, language, voiceover_path)
+            replace_audio_in_video(output_video_path, voiceover_path, output_video_path)
+
+        # Convert translated JSON into a format for the editable table
         editable_table = [
             [float(entry["start"]), entry["original"], entry["translated"], float(entry["end"])]
             for entry in translated_json
@@ -235,10 +262,8 @@ def upload_and_manage(file, language):
         return translated_json, editable_table, output_video_path, elapsed_time_display
 
     except Exception as e:
-        # Handle errors gracefully
         return None, [], None, f"An error occurred: {str(e)}"
 
-
 # Gradio Interface with Tabs
 def build_interface():
     css = """