1. Handling "save_downloaded_files" and "merge_subtitle_with_sources" Options during Transcription Process:

Previously, the "save_downloaded_files" and "merge_subtitle_with_sources" options were processed only right before everything was completed. Now, it has been changed to process after each file is transcribed. When multiple files are input, you can see the processing results more quickly.

2. Aligns the format of saved audio or video file names with subtitle file names when the "save_downloaded_files" or "merge_subtitle_with_sources" option is enabled.

3. Added "input_max_file_name_length" Option: This option adjusts the default maximum length limit for output file names, the default is 100. This option can be configured when launching the app.

app.py

@@ -220,12 +220,41 @@ class WhisperTranscriber:
 
                     # Transcribe
                     result = self.transcribe_file(model, source.source_path, selectedLanguage, task, vadOptions, scaled_progress_listener, **decodeOptions)
-                    filePrefix = slugify(source_prefix + source.get_short_name(), allow_unicode=True)
+                    short_name, suffix = source.get_short_name_suffix(max_length=self.app_config.input_max_file_name_length)
+                    filePrefix = slugify(source_prefix + short_name, allow_unicode=True)
 
                     # Update progress
                     current_progress += source_audio_duration
 
-                    source_download, source_text, source_vtt = self.write_result(result, nllb_model, filePrefix, outputDirectory, highlight_words, scaled_progress_listener)
+                    source_download, source_text, source_vtt = self.write_result(result, nllb_model, filePrefix + suffix.replace(".", "_"), outputDirectory, highlight_words, scaled_progress_listener)
+
+                    if self.app_config.merge_subtitle_with_sources and self.app_config.output_dir is not None:
+                        print("\nmerge subtitle(srt) with source file [" + source.source_name + "]\n")
+                        outRsult = ""
+                        try:
+                            srt_path = source_download[0]
+                            save_path = os.path.join(self.app_config.output_dir, filePrefix)
+                            # save_without_ext, ext = os.path.splitext(save_path)
+                            source_lang = "." + whisper_lang.code if whisper_lang is not None else ""
+                            translate_lang = "." + nllb_lang.code if nllb_lang is not None else ""
+                            output_with_srt = save_path + source_lang + translate_lang + suffix
+        
+                            #ffmpeg -i "input.mp4" -i "input.srt" -c copy -c:s mov_text output.mp4
+                            input_file = ffmpeg.input(source.source_path)
+                            input_srt = ffmpeg.input(srt_path)
+                            out = ffmpeg.output(input_file, input_srt, output_with_srt, vcodec='copy', acodec='copy', scodec='mov_text')
+                            outRsult = out.run(overwrite_output=True)
+                        except Exception as e:
+                            # Ignore error - it's just a cleanup
+                            print("Error merge subtitle with source file: \n" + source.source_path + ", \n" + str(e), outRsult)
+                    elif self.app_config.save_downloaded_files and self.app_config.output_dir is not None and urlData:
+                        print("Saving downloaded file [" + source.source_name + "]")
+                        try:
+                            save_path = os.path.join(self.app_config.output_dir, filePrefix)
+                            shutil.copy(source.source_path, save_path + suffix)
+                        except Exception as e:
+                            # Ignore error - it's just a cleanup
+                            print("Error saving downloaded file: \n" + source.source_path + ", \n" + str(e))
 
                     if len(sources) > 1:
                         # Add new line separators
@@ -272,33 +301,6 @@ class WhisperTranscriber:
                 # Cleanup source
                 if self.deleteUploadedFiles:
                     for source in sources:
-                        if self.app_config.merge_subtitle_with_sources and self.app_config.output_dir is not None and len(source_download) > 0:
-                            print("\nmerge subtitle(srt) with source file [" + source.source_name + "]\n")
-                            outRsult = ""
-                            try:
-                                srt_path = source_download[0]
-                                save_path = os.path.join(self.app_config.output_dir, source.source_name)
-                                save_without_ext, ext = os.path.splitext(save_path)
-                                source_lang = "." + whisper_lang.code if whisper_lang is not None else ""
-                                translate_lang = "." + nllb_lang.code if nllb_lang is not None else ""
-                                output_with_srt = save_without_ext + source_lang + translate_lang + ext
-                                
-                                #ffmpeg -i "input.mp4" -i "input.srt" -c copy -c:s mov_text output.mp4
-                                input_file = ffmpeg.input(source.source_path)
-                                input_srt = ffmpeg.input(srt_path)
-                                out = ffmpeg.output(input_file, input_srt, output_with_srt, vcodec='copy', acodec='copy', scodec='mov_text')
-                                outRsult = out.run(overwrite_output=True)
-                            except Exception as e:
-                                # Ignore error - it's just a cleanup
-                                print("Error merge subtitle with source file: \n" + source.source_path + ", \n" + str(e), outRsult)
-                        elif self.app_config.save_downloaded_files and self.app_config.output_dir is not None and urlData:
-                            print("Saving downloaded file [" + source.source_name + "]")
-                            try:
-                                shutil.copy(source.source_path, self.app_config.output_dir)
-                            except Exception as e:
-                                # Ignore error - it's just a cleanup
-                                print("Error saving downloaded file: \n" + source.source_path + ", \n" + str(e))
-                            
                         print("Deleting temporary source file: " + source.source_path)
                         try:
                             os.remove(source.source_path)
@@ -765,6 +767,8 @@ if __name__ == '__main__':
                         help="True to move downloaded files to outputs directory. This argument will take effect only after output_dir is set.")
     parser.add_argument("--merge_subtitle_with_sources", action='store_true', \
                         help="True to merge subtitle(srt) with sources and move the sources files to the outputs directory. This argument will take effect only after output_dir is set.")
+    parser.add_argument("--input_max_file_name_length", type=int, default=100, \
+                        help="Maximum length of a file name.")
     parser.add_argument("--autolaunch", action='store_true', \
                         help="open the webui URL in the system's default browser upon launch")
                         

src/source.py

@@ -32,10 +32,15 @@ class AudioSource:
         return self.source_name
 
     def get_short_name(self, max_length: int = MAX_FILE_PREFIX_LENGTH):
+        short_name, suffix = self.get_short_name_suffix(max_length=max_length)
+
+        return short_name + suffix
+    
+    def get_short_name_suffix(self, max_length: int = MAX_FILE_PREFIX_LENGTH):
         file_path = pathlib.Path(self.source_name)
-        short_name = file_path.stem[:max_length] + file_path.suffix
+        short_name = file_path.stem[:max_length]
 
-        return short_name
+        return short_name, file_path.suffix
 
     def __str__(self) -> str:
         return self.source_path

src/utils.py

@@ -1,4 +1,4 @@
-import textwrap
+import textwrap
 import unicodedata
 import re
 
@@ -211,7 +211,7 @@ def process_text(text: str, maxLineWidth=None):
     lines = textwrap.wrap(text, width=maxLineWidth, tabsize=4)
     return '\n'.join(lines)
 
-def slugify(value, allow_unicode=False):
+def slugify(value, allow_unicode=False, is_lower=False):
     """
     Taken from https://github.com/django/django/blob/master/django/utils/text.py
     Convert to ASCII if 'allow_unicode' is False. Convert spaces or repeated
@@ -224,7 +224,9 @@ def slugify(value, allow_unicode=False):
         value = unicodedata.normalize('NFKC', value)
     else:
         value = unicodedata.normalize('NFKD', value).encode('ascii', 'ignore').decode('ascii')
-    value = re.sub(r'[^\w\s-]', '', value.lower())
+    if is_lower:
+        value = value.lower()
+    value = re.sub(r'[^\w\s-]', '', value.replace("/","_").replace("⧸","_"))
     return re.sub(r'[-\s]+', '-', value).strip('-_')
 
 def download_file(url: str, destination: str):

src/vadParallel.py

@@ -180,13 +180,13 @@ class ParallelTranscription(AbstractTranscription):
                 total_progress += delta
                 if progress_listener is not None:
                     idx+=1
-                    progress_listener.on_progress(total_progress, total_duration, desc=f"Transcribe parallel: {idx}, {total_progress:.2f}/{total_duration}")
+                    progress_listener.on_progress(total_progress, total_duration, desc=f"Transcribe parallel: {idx}, {total_progress:.2f}/{total_duration:.2f}")
 
             results = results_async.get()
 
             # Call the finished callback
             if progress_listener is not None:
-                progress_listener.on_finished(desc=f"Transcribe parallel: {idx}, {total_progress:.2f}/{total_duration}.")
+                progress_listener.on_finished(desc=f"Transcribe parallel: {idx}, {total_progress:.2f}/{total_duration:.2f}.")
 
             for result in results:
                 # Merge the results