RASMUS commited on
Commit
10419a8
1 Parent(s): db0c1c9

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +9 -8
app.py CHANGED
@@ -117,11 +117,12 @@ def speech_to_text(video_file_path, selected_source_lang, whisper_model):
117
  This space allows you to:
118
  1. Download youtube video with a given url
119
  2. Watch it in the first video component
120
- 3. Run automatic speech recognition on the video using Whisper
121
- 4. Translate the recognized transcriptions to Finnish, Swedish, Danish
122
  5. Burn the translations to the original video and watch the video in the 2nd video component
123
 
124
- Speech Recognition is based on OpenAI Whisper https://github.com/openai/whisper
 
125
  """
126
 
127
  if(video_file_path == None):
@@ -308,14 +309,14 @@ with demo:
308
  ### This space allows you to:
309
  ##### 1. Download youtube video with a given URL
310
  ##### 2. Watch it in the first video component
311
- ##### 3. Run automatic speech recognition on the video using Whisper (Please remember to select translation language)
312
- ##### 4. Translate the recognized transcriptions to Finnish, Swedish, Danish
313
  ##### 5. Burn the translations to the original video and watch the video in the 2nd video component
314
  ''')
315
 
316
  with gr.Column():
317
  gr.Markdown('''
318
- ### 1. Insert Youtube URL below (Some examples below which I suggest to use for first tests)
319
  ##### 1. https://www.youtube.com/watch?v=nlMuHtV82q8&ab_channel=NothingforSale24
320
  ##### 2. https://www.youtube.com/watch?v=JzPfMbG1vrE&ab_channel=ExplainerVideosByLauren
321
  ##### 3. https://www.youtube.com/watch?v=S68vvV0kod8&ab_channel=Pearl-CohnTelevision
@@ -336,7 +337,7 @@ with demo:
336
  with gr.Column():
337
  gr.Markdown('''
338
  ##### Here you can start the transcription and translation process.
339
- ##### Be aware that processing will last for a while (35 second video took around 20 seconds in my testing and might fail for longer videos)
340
  ''')
341
  selected_source_lang.render()
342
  selected_whisper_model.render()
@@ -357,7 +358,7 @@ with demo:
357
  with gr.Column():
358
  gr.Markdown('''
359
  ##### Here you will get translated transcriptions.
360
- ##### Please remember to select Spoken Language and wanted translation language
361
  ##### ''')
362
  selected_translation_lang_2.render()
363
  translate_transcriptions_button = gr.Button("Step 3. Translate transcription")
 
117
  This space allows you to:
118
  1. Download youtube video with a given url
119
  2. Watch it in the first video component
120
+ 3. Run automatic speech recognition on the video using fast Whisper models
121
+ 4. Translate the recognized transcriptions to 26 languages supported by deepL
122
  5. Burn the translations to the original video and watch the video in the 2nd video component
123
 
124
+ Speech Recognition is based on models from OpenAI Whisper https://github.com/openai/whisper
125
+ This space is using c++ implementation by https://github.com/ggerganov/whisper.cpp
126
  """
127
 
128
  if(video_file_path == None):
 
309
  ### This space allows you to:
310
  ##### 1. Download youtube video with a given URL
311
  ##### 2. Watch it in the first video component
312
+ ##### 3. Run automatic speech recognition on the video using Whisper
313
+ ##### 4. Translate the recognized transcriptions to 26 languages supported by deepL
314
  ##### 5. Burn the translations to the original video and watch the video in the 2nd video component
315
  ''')
316
 
317
  with gr.Column():
318
  gr.Markdown('''
319
+ ### 1. Insert Youtube URL below. Some test videos below:
320
  ##### 1. https://www.youtube.com/watch?v=nlMuHtV82q8&ab_channel=NothingforSale24
321
  ##### 2. https://www.youtube.com/watch?v=JzPfMbG1vrE&ab_channel=ExplainerVideosByLauren
322
  ##### 3. https://www.youtube.com/watch?v=S68vvV0kod8&ab_channel=Pearl-CohnTelevision
 
337
  with gr.Column():
338
  gr.Markdown('''
339
  ##### Here you can start the transcription and translation process.
340
+ ##### Be aware that processing will last some time. With base model it is around 3x speed
341
  ''')
342
  selected_source_lang.render()
343
  selected_whisper_model.render()
 
358
  with gr.Column():
359
  gr.Markdown('''
360
  ##### Here you will get translated transcriptions.
361
+ ##### Please remember to select target language
362
  ##### ''')
363
  selected_translation_lang_2.render()
364
  translate_transcriptions_button = gr.Button("Step 3. Translate transcription")