Update app.py

app.py

@@ -32,7 +32,7 @@ print("cwd", os.getcwd())
 print(os.listdir())
 
         
-def process_video(Video, target_language):
+def process_video(Video, target_language, translate_video):
     current_path = os.getcwd()
     print("Iniciando process_video")
     common_uuid = uuid.uuid4()
@@ -40,16 +40,13 @@ def process_video(Video, target_language):
     run(["ffmpeg", "-version"])
     audio_file = f"{common_uuid}.wav"
     run(["ffmpeg", "-i", Video, audio_file])
-    transcript_file = f"{common_uuid}.srt"
+    transcript_file = f"{current_path}/{common_uuid}.srt"
+
     # Transcription with Whisper.
     print("Iniciando transcrição com Whisper")
     segments, _ = whisper_model.transcribe(audio_file, beam_size=5)
     segments = list(segments)
-    transcript_file = f"{current_path}/{common_uuid}.srt"
-    
-    # Create a list to hold the translated lines.
-    translated_lines = []
-    
+
     with open(transcript_file, "w+", encoding="utf-8") as f:
         counter = 1
         for segment in segments:
@@ -70,28 +67,25 @@ def process_video(Video, target_language):
             f.write(f"{formatted_start} --> {formatted_end}\n")
             f.write(f"{segment.text}\n\n")
             counter += 1
-        # Move the file pointer to the beginning of the file.
-        f.seek(0)
-    
-        # Translating the SRT from Whisper with Google Translate.
-        target_language_code = google_lang_codes.get(target_language, "en")
-        paragraph = ""
-        for line in f:
-            if line.strip().isnumeric() or "-->" in line:
-                translated_lines.append(line)
-            elif line.strip() != "":
-                translated_text = translator.translate(line.strip(), dest=target_language_code).text
-                translated_lines.append(translated_text + "\n")
-            else:
-                translated_lines.append("\n")
-    
-        # Move the file pointer to the beginning of the file and truncate it.
-        f.seek(0)
-        f.truncate()
-    
-        # Write the translated lines back into the original file.
-        f.writelines(translated_lines)
-    #return None, None
+
+        # Check if translation is needed
+        if translate_video:
+            # Translating the SRT from Whisper with Google Translate.
+            target_language_code = google_lang_codes.get(target_language, "en")
+            translated_lines = []
+            f.seek(0)  # Move the file pointer to the beginning of the file.
+            for line in f:
+                if line.strip().isnumeric() or "-->" in line:
+                    translated_lines.append(line)
+                elif line.strip() != "":
+                    translated_text = translator.translate(line.strip(), dest=target_language_code).text
+                    translated_lines.append(translated_text + "\n")
+                else:
+                    translated_lines.append("\n")
+            
+            f.seek(0)  # Move the file pointer to the beginning of the file and truncate it.
+            f.truncate()
+            f.writelines(translated_lines)  # Write the translated lines back into the original file.
     output_video = f"{common_uuid}_output_video.mp4"
     # Debugging: Validate FFmpeg command for subtitle embedding
     print("Validating FFmpeg command for subtitle embedding...")
@@ -128,6 +122,7 @@ iface = gr.Interface(
     inputs=[
         gr.Video(),
         gr.Dropdown(choices=list(google_lang_codes.keys()), label="Target Language for Translation", value="English"),
+        gr.Checkbox(label="Translate Video", value=True, info="Check to translate the video to the selected language. Uncheck for transcription only."),
     ],
     outputs=[
         gr.Video(),