Hugging Face's logo

Spaces:
Ritesh-hf
/
speech-to-text-with-timestamps
Sleeping

App Files Community
Ritesh-hf commited on
Commit
45a4204
·
verified ·
1 Parent(s): 078e192

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +18 -18
app.py CHANGED
@@ -126,11 +126,11 @@ def transcribe(input_audio, audio_language, task, timestamp_type):
126
  return [text, timestamps, foul_words, (sample_rate, audio_data)]
127
 
128
  examples = [
129
- #["arabic_english_audios/audios/arabic_audio_1.mp3", 'Arabic', 'transcribe', 'word'],
130
- #["arabic_english_audios/audios/arabic_audio_2.flac", 'Arabic', 'transcribe', 'word'],
131
- #["arabic_english_audios/audios/arabic_audio_3.flac", 'Arabic', 'transcribe', 'word'],
132
- #["arabic_english_audios/audios/arabic_audio_31.mp3", 'Arabic', 'transcribe', 'word'],
133
- #["arabic_english_audios/audios/arabic_audio_32.mp3", 'Arabic', 'transcribe', 'word'],
134
 
135
  ["arabic_english_audios/audios/english_audio_31.mp3", 'English', 'transcribe', 'word'],
136
  ["arabic_english_audios/audios/english_audio_32.mp3", 'English', 'transcribe', 'word'],
@@ -138,18 +138,18 @@ examples = [
138
  ["arabic_english_audios/audios/english_audio_34.mp3", 'English', 'transcribe', 'word'],
139
  ["arabic_english_audios/audios/english_audio_35.mp3", 'English', 'transcribe', 'word'],
140
  ["arabic_english_audios/audios/english_audio_36.mp3", 'English', 'transcribe', 'word'],
141
- # ["arabic_english_audios/audios/english_audio_37.mp3", 'English', 'transcribe', 'word'],
142
- # ["arabic_english_audios/audios/english_audio_38.mp3", 'English', 'transcribe', 'word'],
143
- # ["arabic_english_audios/audios/english_audio_39.mp3", 'English', 'transcribe', 'word'],
144
- # ["arabic_english_audios/audios/english_audio_40.mp3", 'English', 'transcribe', 'word'],
145
- # ["arabic_english_audios/audios/english_audio_41.mp3", 'English', 'transcribe', 'word'],
146
- # ["arabic_english_audios/audios/english_audio_42.mp3", 'English', 'transcribe', 'word'],
147
- # ["arabic_english_audios/audios/english_audio_43.mp3", 'English', 'transcribe', 'word'],
148
- # ["arabic_english_audios/audios/english_audio_44.mp3", 'English', 'transcribe', 'word'],
149
- # ["arabic_english_audios/audios/english_audio_45.mp3", 'English', 'transcribe', 'word'],
150
- # ["arabic_english_audios/audios/english_audio_46.mp3", 'English', 'transcribe', 'word'],
151
- # ["arabic_english_audios/audios/english_audio_48.mp3", 'English', 'transcribe', 'word'],
152
- # ["arabic_english_audios/audios/english_audio_49.mp3", 'English', 'transcribe', 'word'],
153
  ]
154
 
155
  with gr.Blocks(theme=gr.themes.Default()) as demo:
@@ -171,7 +171,7 @@ with gr.Blocks(theme=gr.themes.Default()) as demo:
171
  foul_words = gr.Text(label="Foul Words")
172
  output_audio = gr.Audio(label="Output Audio", type="numpy")
173
 
174
- examples = gr.Examples(examples, inputs=[audio_input, audio_language, task, timestamp_type], outputs=[transcript_output, timestamp_output, foul_words, output_audio], fn=transcribe, examples_per_page=20)
175
 
176
  submit_button.click(fn=transcribe, inputs=[audio_input, audio_language, task, timestamp_type], outputs=[transcript_output, timestamp_output, foul_words, output_audio])
177
  clear_button.add([audio_input, audio_language, task, timestamp_type, transcript_output, timestamp_output, foul_words, output_audio])
 
126
  return [text, timestamps, foul_words, (sample_rate, audio_data)]
127
 
128
  examples = [
129
+ ["arabic_english_audios/audios/arabic_audio_1.mp3", 'Arabic', 'transcribe', 'word'],
130
+ ["arabic_english_audios/audios/arabic_audio_2.flac", 'Arabic', 'transcribe', 'word'],
131
+ ["arabic_english_audios/audios/arabic_audio_3.flac", 'Arabic', 'transcribe', 'word'],
132
+ ["arabic_english_audios/audios/arabic_audio_31.mp3", 'Arabic', 'transcribe', 'word'],
133
+ ["arabic_english_audios/audios/arabic_audio_32.mp3", 'Arabic', 'transcribe', 'word'],
134
 
135
  ["arabic_english_audios/audios/english_audio_31.mp3", 'English', 'transcribe', 'word'],
136
  ["arabic_english_audios/audios/english_audio_32.mp3", 'English', 'transcribe', 'word'],
 
138
  ["arabic_english_audios/audios/english_audio_34.mp3", 'English', 'transcribe', 'word'],
139
  ["arabic_english_audios/audios/english_audio_35.mp3", 'English', 'transcribe', 'word'],
140
  ["arabic_english_audios/audios/english_audio_36.mp3", 'English', 'transcribe', 'word'],
141
+ ["arabic_english_audios/audios/english_audio_37.mp3", 'English', 'transcribe', 'word'],
142
+ ["arabic_english_audios/audios/english_audio_38.mp3", 'English', 'transcribe', 'word'],
143
+ ["arabic_english_audios/audios/english_audio_39.mp3", 'English', 'transcribe', 'word'],
144
+ ["arabic_english_audios/audios/english_audio_40.mp3", 'English', 'transcribe', 'word'],
145
+ ["arabic_english_audios/audios/english_audio_41.mp3", 'English', 'transcribe', 'word'],
146
+ ["arabic_english_audios/audios/english_audio_42.mp3", 'English', 'transcribe', 'word'],
147
+ ["arabic_english_audios/audios/english_audio_43.mp3", 'English', 'transcribe', 'word'],
148
+ ["arabic_english_audios/audios/english_audio_44.mp3", 'English', 'transcribe', 'word'],
149
+ ["arabic_english_audios/audios/english_audio_45.mp3", 'English', 'transcribe', 'word'],
150
+ ["arabic_english_audios/audios/english_audio_46.mp3", 'English', 'transcribe', 'word'],
151
+ ["arabic_english_audios/audios/english_audio_48.mp3", 'English', 'transcribe', 'word'],
152
+ ["arabic_english_audios/audios/english_audio_49.mp3", 'English', 'transcribe', 'word'],
153
  ]
154
 
155
  with gr.Blocks(theme=gr.themes.Default()) as demo:
 
171
  foul_words = gr.Text(label="Foul Words")
172
  output_audio = gr.Audio(label="Output Audio", type="numpy")
173
 
174
+ examples = gr.Examples(examples, inputs=[audio_input, audio_language, task, timestamp_type], outputs=[transcript_output, timestamp_output, foul_words, output_audio], fn=transcribe, examples_per_page=20, cache_examples=False)
175
 
176
  submit_button.click(fn=transcribe, inputs=[audio_input, audio_language, task, timestamp_type], outputs=[transcript_output, timestamp_output, foul_words, output_audio])
177
  clear_button.add([audio_input, audio_language, task, timestamp_type, transcript_output, timestamp_output, foul_words, output_audio])