renatotn7 commited on
Commit
7f51198
·
1 Parent(s): 3892f6c

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +7 -16
app.py CHANGED
@@ -1,5 +1,5 @@
1
  from typing import Iterator
2
- import streamlit as st
3
  from io import StringIO
4
  import os
5
  import pathlib
@@ -17,7 +17,7 @@ from src.utils import slugify, write_srt, write_vtt
17
  from src.vad import NonSpeechStrategy, PeriodicTranscriptionConfig, TranscriptionConfig, VadPeriodicTranscription, VadSileroTranscription
18
 
19
  # Limitations (set to -1 to disable)
20
- DEFAULT_INPUT_AUDIO_MAX_DURATION = 600 # seconds
21
 
22
  # Whether or not to automatically delete all uploaded files, to save disk space
23
  DELETE_UPLOADED_FILES = True
@@ -221,10 +221,11 @@ def create_ui(inputAudioMaxDuration, share=False, server_name: str = None):
221
  ui_article = "Read the [documentation here](https://huggingface.co/spaces/aadnk/whisper-webui/blob/main/docs/options.md)"
222
 
223
  demo = gr.Interface(fn=ui.transcribe_webui, description=ui_description, article=ui_article, inputs=[
224
- gr.Dropdown(choices=["tiny", "base", "small", "medium", "large"], value="medium", label="Model"),
225
  gr.Dropdown(choices=sorted(LANGUAGES), label="Language"),
226
  gr.Text(label="URL (YouTube, etc.)"),
227
- gr.Audio(source="upload", type="filepath", label="Upload Audio"),
 
228
  gr.Audio(source="microphone", type="filepath", label="Microphone Input"),
229
  gr.Dropdown(choices=["transcribe", "translate"], label="Task"),
230
  gr.Dropdown(choices=["none", "silero-vad", "silero-vad-skip-gaps", "silero-vad-expand-into-gaps", "periodic-vad"], label="VAD"),
@@ -236,21 +237,11 @@ def create_ui(inputAudioMaxDuration, share=False, server_name: str = None):
236
  gr.File(label="Download"),
237
  gr.Text(label="Transcription"),
238
  gr.Text(label="Segments")
 
239
  ])
240
 
241
  demo.launch(share=share, server_name=server_name)
242
 
243
  if __name__ == '__main__':
244
- uploaded_file = st.file_uploader("Choose a file")
245
- if uploaded_file is not None:
246
- # To read file as bytes:
247
- bytes_data = uploaded_file.getvalue()
248
- st.video(bytes_data)
249
- os.system("rm -R /tmp/*")
250
- os.system("mkdir /tmp/_input")
251
- os.system("mkdir /tmp/_output")
252
- os.system("mkdir /tmp/_outputf")
253
- os.system("mkdir /tmp/videoSaida")
254
- with open("/tmp/inputvideo", "wb") as binary_file:
255
- binary_file.write(bytes_data)
256
  create_ui(DEFAULT_INPUT_AUDIO_MAX_DURATION)
 
1
  from typing import Iterator
2
+
3
  from io import StringIO
4
  import os
5
  import pathlib
 
17
  from src.vad import NonSpeechStrategy, PeriodicTranscriptionConfig, TranscriptionConfig, VadPeriodicTranscription, VadSileroTranscription
18
 
19
  # Limitations (set to -1 to disable)
20
+ DEFAULT_INPUT_AUDIO_MAX_DURATION = 1200 # seconds
21
 
22
  # Whether or not to automatically delete all uploaded files, to save disk space
23
  DELETE_UPLOADED_FILES = True
 
221
  ui_article = "Read the [documentation here](https://huggingface.co/spaces/aadnk/whisper-webui/blob/main/docs/options.md)"
222
 
223
  demo = gr.Interface(fn=ui.transcribe_webui, description=ui_description, article=ui_article, inputs=[
224
+ gr.Dropdown(choices=["medium.en", "large"], value="medium.en", label="Model"),
225
  gr.Dropdown(choices=sorted(LANGUAGES), label="Language"),
226
  gr.Text(label="URL (YouTube, etc.)"),
227
+ gr.Audio(source="upload", type="filepath", label="Upload Audio"),
228
+ gr.File(label="submeta seu video",interactive=True)
229
  gr.Audio(source="microphone", type="filepath", label="Microphone Input"),
230
  gr.Dropdown(choices=["transcribe", "translate"], label="Task"),
231
  gr.Dropdown(choices=["none", "silero-vad", "silero-vad-skip-gaps", "silero-vad-expand-into-gaps", "periodic-vad"], label="VAD"),
 
237
  gr.File(label="Download"),
238
  gr.Text(label="Transcription"),
239
  gr.Text(label="Segments")
240
+ gr.Text(label="Videos")
241
  ])
242
 
243
  demo.launch(share=share, server_name=server_name)
244
 
245
  if __name__ == '__main__':
246
+
 
 
 
 
 
 
 
 
 
 
 
247
  create_ui(DEFAULT_INPUT_AUDIO_MAX_DURATION)