marquesafonso commited on
Commit
f13603f
·
verified ·
1 Parent(s): 8562af7

add tab selection

Browse files
Files changed (1) hide show
  1. app.py +30 -30
app.py CHANGED
@@ -5,36 +5,36 @@ def main():
5
  with gr.Blocks(title='multilang-asr-transcriber', delete_cache=(86400, 86400), theme=gr.themes.Base()) as demo:
6
  gr.Markdown('## Multilang ASR Transcriber')
7
  gr.Markdown('An automatic speech recognition tool using [faster-whisper](https://github.com/SYSTRAN/faster-whisper). Supports multilingual video transcription and translation to english. Users may set the max words per line.')
8
-
9
- with gr.Tab("Video"):
10
- video = True
11
- file = gr.File(file_types=["video"],type="filepath", label="Upload a video")
12
- file_type = gr.Radio(choices=["video"], value="video", label="File Type")
13
- max_words_per_line = gr.Number(value=6, label="Max words per line")
14
- task = gr.Radio(choices=["transcribe", "translate"], value="transcribe", label="Select Task")
15
- model_version = gr.Radio(choices=["deepdml/faster-whisper-large-v3-turbo-ct2", "large-v3"], value="deepdml/faster-whisper-large-v3-turbo-ct2", label="Select Model")
16
- text_output = gr.Textbox(label="SRT Text transcription", show_copy_button=True)
17
- srt_file = gr.File(file_count="single", type="filepath", file_types=[".srt"], label="SRT file")
18
- text_clean_output = gr.Textbox(label="Text transcription", show_copy_button=True)
19
- gr.Interface(transcriber,
20
- inputs=[file, file_type, max_words_per_line, task, model_version],
21
- outputs=[text_output, srt_file, text_clean_output],
22
- allow_flagging="never")
23
-
24
- with gr.Tab("Audio"):
25
- video = False
26
- file = gr.File(file_types=["audio"],type="filepath", label="Upload an audio file")
27
- file_type = gr.Radio(choices=["audio"], value="audio", label="File Type")
28
- max_words_per_line = gr.Number(value=6, label="Max words per line")
29
- task = gr.Radio(choices=["transcribe", "translate"], value="transcribe", label="Select Task")
30
- model_version = gr.Radio(choices=["deepdml/faster-whisper-large-v3-turbo-ct2", "large-v3"], value="deepdml/faster-whisper-large-v3-turbo-ct2", label="Select Model")
31
- text_output = gr.Textbox(label="SRT Text transcription", show_copy_button=True)
32
- srt_file = gr.File(file_count="single", type="filepath", file_types=[".srt"], label="SRT file")
33
- text_clean_output = gr.Textbox(label="Text transcription", show_copy_button=True)
34
- gr.Interface(transcriber,
35
- inputs=[file, file_type, max_words_per_line, task, model_version],
36
- outputs=[text_output, srt_file, text_clean_output],
37
- allow_flagging="never")
38
  demo.launch()
39
 
40
  if __name__ == '__main__':
 
5
  with gr.Blocks(title='multilang-asr-transcriber', delete_cache=(86400, 86400), theme=gr.themes.Base()) as demo:
6
  gr.Markdown('## Multilang ASR Transcriber')
7
  gr.Markdown('An automatic speech recognition tool using [faster-whisper](https://github.com/SYSTRAN/faster-whisper). Supports multilingual video transcription and translation to english. Users may set the max words per line.')
8
+ with gr.Tabs(selected="video") as tabs:
9
+ with gr.Tab("Video", id="video"):
10
+ video = True
11
+ file = gr.File(file_types=["video"],type="filepath", label="Upload a video")
12
+ file_type = gr.Radio(choices=["video"], value="video", label="File Type")
13
+ max_words_per_line = gr.Number(value=6, label="Max words per line")
14
+ task = gr.Radio(choices=["transcribe", "translate"], value="transcribe", label="Select Task")
15
+ model_version = gr.Radio(choices=["deepdml/faster-whisper-large-v3-turbo-ct2", "large-v3"], value="deepdml/faster-whisper-large-v3-turbo-ct2", label="Select Model")
16
+ text_output = gr.Textbox(label="SRT Text transcription", show_copy_button=True)
17
+ srt_file = gr.File(file_count="single", type="filepath", file_types=[".srt"], label="SRT file")
18
+ text_clean_output = gr.Textbox(label="Text transcription", show_copy_button=True)
19
+ gr.Interface(transcriber,
20
+ inputs=[file, file_type, max_words_per_line, task, model_version],
21
+ outputs=[text_output, srt_file, text_clean_output],
22
+ allow_flagging="never")
23
+
24
+ with gr.Tab("Audio", id = "audio"):
25
+ video = False
26
+ file = gr.File(file_types=["audio"],type="filepath", label="Upload an audio file")
27
+ file_type = gr.Radio(choices=["audio"], value="audio", label="File Type")
28
+ max_words_per_line = gr.Number(value=6, label="Max words per line")
29
+ task = gr.Radio(choices=["transcribe", "translate"], value="transcribe", label="Select Task")
30
+ model_version = gr.Radio(choices=["deepdml/faster-whisper-large-v3-turbo-ct2", "large-v3"], value="deepdml/faster-whisper-large-v3-turbo-ct2", label="Select Model")
31
+ text_output = gr.Textbox(label="SRT Text transcription", show_copy_button=True)
32
+ srt_file = gr.File(file_count="single", type="filepath", file_types=[".srt"], label="SRT file")
33
+ text_clean_output = gr.Textbox(label="Text transcription", show_copy_button=True)
34
+ gr.Interface(transcriber,
35
+ inputs=[file, file_type, max_words_per_line, task, model_version],
36
+ outputs=[text_output, srt_file, text_clean_output],
37
+ allow_flagging="never")
38
  demo.launch()
39
 
40
  if __name__ == '__main__':