akadriu commited on
Commit
a878076
·
verified ·
1 Parent(s): 192be9f

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +6 -5
app.py CHANGED
@@ -1,7 +1,6 @@
1
  import os
2
  from transformers import pipeline
3
  import gradio as gr
4
- import numpy as np
5
 
6
  # Fetch the token from the environment
7
  hf_token = os.getenv("HUGGINGFACE_HUB_TOKEN")
@@ -9,10 +8,11 @@ model_id = "akadriu/whisper-medium-sq" # update with your model id
9
  pipe = pipeline("automatic-speech-recognition", model=model_id, token=hf_token)
10
 
11
  def transcribe_speech(filepath):
12
- # Load the audio file into a numpy array
13
  if filepath is None:
14
  raise ValueError("No audio file provided.")
15
 
 
16
  output = pipe(
17
  filepath,
18
  max_new_tokens=256,
@@ -25,16 +25,16 @@ def transcribe_speech(filepath):
25
  )
26
  return output["text"]
27
 
28
- # Create Gradio interface
29
  mic_transcribe = gr.Interface(
30
  fn=transcribe_speech,
31
- inputs=gr.Audio(source="microphone", type="filepath"), # Removed plural from "sources"
32
  outputs="text",
33
  )
34
 
35
  file_transcribe = gr.Interface(
36
  fn=transcribe_speech,
37
- inputs=gr.Audio(source="upload", type="filepath"), # Removed plural from "sources"
38
  outputs="text",
39
  )
40
 
@@ -48,3 +48,4 @@ with demo:
48
 
49
  demo.launch(debug=True)
50
 
 
 
1
  import os
2
  from transformers import pipeline
3
  import gradio as gr
 
4
 
5
  # Fetch the token from the environment
6
  hf_token = os.getenv("HUGGINGFACE_HUB_TOKEN")
 
8
  pipe = pipeline("automatic-speech-recognition", model=model_id, token=hf_token)
9
 
10
  def transcribe_speech(filepath):
11
+ # Check if the filepath is valid
12
  if filepath is None:
13
  raise ValueError("No audio file provided.")
14
 
15
+ # Perform speech transcription
16
  output = pipe(
17
  filepath,
18
  max_new_tokens=256,
 
25
  )
26
  return output["text"]
27
 
28
+ # Create Gradio interfaces without the 'source' argument
29
  mic_transcribe = gr.Interface(
30
  fn=transcribe_speech,
31
+ inputs=gr.Audio(type="filepath"),
32
  outputs="text",
33
  )
34
 
35
  file_transcribe = gr.Interface(
36
  fn=transcribe_speech,
37
+ inputs=gr.Audio(type="filepath"),
38
  outputs="text",
39
  )
40
 
 
48
 
49
  demo.launch(debug=True)
50
 
51
+