minor fixes

adjust.py

@@ -101,31 +101,32 @@ def get_speakers_previous(to_name):
     return get_current()
 
 
-def start_adjust(input_file, to_name, progress=gr.Progress()):
+def start_adjust(input_file, to_name, start_time, end_time, progress=gr.Progress()):
     change_name(to_name)
 
     # Replacing texts
     progress(0.4, desc=ui_lang["progress_adjust_speaker"])
-    transcribe_txt_list, subtitle_txt_list = utils.read_transcribe_subtitle_file(
-        False)
+    transcribe_txt_list, subtitle_txt_list = utils.read_transcribe_subtitle_file(input_file,
+                                                                                 False)
     modified_transcribe = replace_text(transcribe_txt_list)
     modified_subtitle = replace_text(subtitle_txt_list)
     utils.write_transcribe_subtitle_file(
-        modified_transcribe, modified_subtitle, True)
+        input_file, modified_transcribe, modified_subtitle, True)
 
     # Adding subtitle to video
     progress(0.8, desc=ui_lang["progress_add_subtitle"])
     add_subtitle_to_video(
-        input_file, get_subtitle_adjusted_file(), get_video_subtitle_file(), start_time_for_adjustment, end_time_for_adjustment)
+        input_file, subtitle_adjusted_file, video_subtitle_file, start_time, end_time)
 
     # Return video file link, transcribe string, transcribe.txt, subtitle.txt
-    transcribe_txt_list, _ = utils.read_transcribe_subtitle_file(True)
+    transcribe_txt_list, _ = utils.read_transcribe_subtitle_file(
+        input_file, True)
     print(line for line in transcribe_txt_list)
     transcribe_txt = "\n".join(transcribe_txt_list)
     return [
-        get_video_subtitle_file(),
+        video_subtitle_file,
         transcribe_txt,
-        [get_transcribe_adjusted_file(), get_subtitle_adjusted_file()]
+        [transcribe_adjusted_file, subtitle_adjusted_file]
     ]
 
 

app.py

@@ -52,7 +52,7 @@ with gr.Blocks() as demo:
                           adjust_speaker, adjust_audio])
 
     adjust_button.render()
-    adjust_button.click(start_adjust, inputs=[input_video, adjust_speaker], outputs=[
+    adjust_button.click(start_adjust, inputs=[input_video, adjust_speaker, start_time, end_time,], outputs=[
                         output_video, output_transcribe, output_file])
 
     with gr.Accordion("Copyright"):

diarization.py

@@ -36,7 +36,7 @@ def start_diarization(input_file):
         print(
             f"start={turn.start:.3f}s stop={turn.end:.3f}s speaker_{speaker}")
 
-    save_groups_json(sample_groups, speaker_groups)
+    save_groups_json(input_file, sample_groups, speaker_groups)
     audio_segmentation(input_file, speaker_groups)
     print(str(speaker_groups))
     return str(speaker_groups)
@@ -51,8 +51,8 @@ def audio_segmentation(input_file, speaker_groups_dict):
         print(f"group {speaker}: {time[0]*1000}--{time[1]*1000}")
 
 
-def save_groups_json(sample_groups_list: list, speaker_groups_dict: dict):
-    with open(get_sample_groups_json(), "w", encoding="utf-8") as json_file_sample:
+def save_groups_json(input_file, sample_groups_list: list, speaker_groups_dict: dict):
+    with open(sample_groups_json, "w", encoding="utf-8") as json_file_sample:
         json.dump(sample_groups_list, json_file_sample)
-    with open(get_speaker_groups_json(), "w", encoding="utf-8") as json_file_speaker:
+    with open(speaker_groups_json, "w", encoding="utf-8") as json_file_speaker:
         json.dump(speaker_groups_dict, json_file_speaker)

file_name.py

@@ -3,44 +3,11 @@ import os
 
 current_working_directory = os.getcwd()
 
-current_input_file = ""
-
-
-def get_title():
-    return os.path.basename(current_input_file)
-
-
-def get_audio_file():
-    return f"{current_working_directory}/input_{get_title()}.wav"
-
-
-def get_transcribe_file():
-    return f"{current_working_directory}/transcribe_{get_title()}.txt"
-
-
-def get_subtitle_file():
-    return f"{current_working_directory}/subtitle_{get_title()}.srt"
-
-
-def get_transcribe_adjusted_file():
-    return f"{current_working_directory}/transcribe_adjusted_{get_title()}.txt"
-
-
-def get_subtitle_adjusted_file():
-    return f"{current_working_directory}/subtitle_adjusted_{get_title()}.srt"
-
-
-def get_video_subtitle_file():
-    return f"{current_working_directory}/output_{get_title()}.mp4"
-
-
-def get_sample_groups_json():
-    return f"{current_working_directory}/sample_groups_{get_title()}.json"
-
-
-def get_speaker_groups_json():
-    return f"{current_working_directory}/speaker_groups_{get_title()}.json"
-
-
-start_time_for_adjustment = "00:00:00"
-end_time_for_adjustment = "01:00:00"
+audio_file = f"{current_working_directory}/input.wav"
+transcribe_file = f"{current_working_directory}/transcribe.txt"
+subtitle_file = f"{current_working_directory}/subtitle.srt"
+transcribe_adjusted_file = f"{current_working_directory}/transcribe_adjusted.txt"
+subtitle_adjusted_file = f"{current_working_directory}/subtitle_adjusted.srt"
+video_subtitle_file = f"{current_working_directory}/output.mp4"
+sample_groups_json = f"{current_working_directory}/sample_groups.json"
+speaker_groups_json = f"{current_working_directory}/speaker_groups.json"

set_up.py

@@ -28,42 +28,35 @@ def prepare_input(input_file, start_time, end_time, lang, model_size, use_summar
         gr.Warning(ui_lang["model_dropdown_warning"])
         return [None, None, [None, None]]
 
-    global current_input_file
-    current_input_file = input_file
-    global start_time_for_adjustment
-    start_time_for_adjustment = start_time
-    global end_time_for_adjustment
-    end_time_for_adjustment = end_time
-
     print(f"SOURCE: {input_file}")
-    print(f"VIDEO FILE: {current_input_file}")
-    print(f"AUDIO FILE: {get_audio_file()}")
+    print(f"AUDIO FILE: {audio_file}")
 
     # Convert video to audio
     progress(0.2, desc=ui_lang["progress_preparing_video"])
     convert_video_to_audio(
-        input_file, get_audio_file(), start_time, end_time)
+        input_file, audio_file, start_time, end_time)
 
     # Start diarization
     progress(0.4, desc=ui_lang["progress_acquiring_diarization"])
-    start_diarization(get_audio_file())
+    start_diarization(audio_file)
 
     # Start transcribing
     progress(0.6, desc=ui_lang["progress_transcribing_audio"])
-    start_transcribe(lang, model_size, progress)
+    start_transcribe(input_file, lang, model_size, progress)
 
     # Add subtitle to video
     progress(0.8, desc=ui_lang["progress_add_subtitle"])
-    add_subtitle_to_video(input_file, get_subtitle_file(),
-                          get_video_subtitle_file(), start_time, end_time)
+    add_subtitle_to_video(input_file, subtitle_file,
+                          video_subtitle_file, start_time, end_time)
 
     # Return video file link, transcribe string, transcribe.txt, subtitle.txt
-    transcribe_txt_list, _ = utils.read_transcribe_subtitle_file(False)
+    transcribe_txt_list, _ = utils.read_transcribe_subtitle_file(
+        input_file, False)
     transcribe_txt = "\n".join(transcribe_txt_list)
     return [
-        get_video_subtitle_file(),
+        video_subtitle_file,
         transcribe_txt,
-        [get_transcribe_file(), get_subtitle_file()]
+        [transcribe_file, subtitle_file]
     ]
 
 

transcribe.py

@@ -12,7 +12,7 @@ model_lang_list = ['en', 'id', None]
 model_size = ["tiny", "base", "small", "medium", "large"]
 
 
-def start_transcribe(lang_choice: int, model_size_choice: int, progress):
+def start_transcribe(input_file, lang_choice: int, model_size_choice: int, progress):
 
     print(
         f"Starting transcribing with model size {model_size[model_size_choice]} for language {model_lang_list[lang_choice]}")
@@ -49,7 +49,7 @@ def start_transcribe(lang_choice: int, model_size_choice: int, progress):
         transcribe_txt_list.append(f"[{name}] {speaker_txt}\n")
 
     utils.write_transcribe_subtitle_file(
-        transcribe_txt_list, subtitle_txt_list, False)
+        input_file, transcribe_txt_list, subtitle_txt_list, False)
 
 
 def time_str(t):

utils.py

@@ -4,23 +4,23 @@ import os
 
 
 def load_groups_json():
-    if not os.path.exists(get_sample_groups_json()) or not os.path.exists(get_speaker_groups_json()):
+    if not os.path.exists(sample_groups_json) or not os.path.exists(speaker_groups_json):
         print("JSON file doesn't exist")
         return [], {}
 
-    with open(get_sample_groups_json(), "r", encoding="utf-8") as json_file_sample:
+    with open(sample_groups_json, "r", encoding="utf-8") as json_file_sample:
         sample_groups_list: list = json.load(json_file_sample)
-    with open(get_speaker_groups_json(), "r", encoding="utf-8") as json_file_speaker:
+    with open(speaker_groups_json, "r", encoding="utf-8") as json_file_speaker:
         speaker_groups_dict: dict = json.load(json_file_speaker)
     return sample_groups_list, speaker_groups_dict
 
 
-def write_transcribe_subtitle_file(transcribe_txt_list: list, subtitle_txt_list: list, adjustment: bool):
-    transcribe = get_transcribe_file()
-    subtitle = get_subtitle_file()
+def write_transcribe_subtitle_file(input_file, transcribe_txt_list: list, subtitle_txt_list: list, adjustment: bool):
+    transcribe = transcribe_file
+    subtitle = subtitle_file
     if adjustment:
-        transcribe = get_transcribe_adjusted_file()
-        subtitle = get_subtitle_adjusted_file()
+        transcribe = transcribe_adjusted_file
+        subtitle = subtitle_adjusted_file
 
     with open(transcribe, "w", encoding="utf-8") as file:
         file.writelines(transcribe_txt_list)
@@ -28,12 +28,12 @@ def write_transcribe_subtitle_file(transcribe_txt_list: list, subtitle_txt_list:
         file.writelines(subtitle_txt_list)
 
 
-def read_transcribe_subtitle_file(adjustment: bool):
-    transcribe = get_transcribe_file()
-    subtitle = get_subtitle_file()
+def read_transcribe_subtitle_file(input_file, adjustment: bool):
+    transcribe = transcribe_file
+    subtitle = subtitle_file
     if adjustment:
-        transcribe = get_transcribe_adjusted_file()
-        subtitle = get_subtitle_adjusted_file()
+        transcribe = transcribe_adjusted_file
+        subtitle = subtitle_adjusted_file
 
     if not os.path.exists(transcribe) or not os.path.exists(subtitle):
         print("Transcribe or subtitle file doesn't exist")