Mark0047 commited on
Commit
90b4970
·
verified ·
1 Parent(s): 7a94e97

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +17 -15
app.py CHANGED
@@ -1,21 +1,23 @@
 
 
 
 
 
1
  import gradio as gr
2
 
3
- # Define a processing function that directly returns the transcription
4
  def process_transcription(audio_input):
5
- # Load the Whisper model and get the result
6
- model = gr.load("models/openai/whisper-large-v3-turbo") # Use gr.load in the correct way
7
  result = model(audio_input)
8
 
9
- # Return only the transcription text
10
- return result["text"]
11
-
12
- # Use a Gradio interface to launch the app
13
- with gr.Blocks() as demo:
14
- audio_input = gr.Audio(source="microphone", type="filepath", label="Input Audio")
15
- transcription_output = gr.Textbox(label="Transcription")
16
-
17
- # Add interaction
18
- audio_input.change(process_transcription, inputs=audio_input, outputs=transcription_output)
19
 
20
- # Launch the app
21
- demo.launch()
 
 
 
 
 
 
1
+ # import gradio as gr
2
+
3
+ # gr.load("models/openai/whisper-large-v3-turbo").launch()
4
+
5
+
6
  import gradio as gr
7
 
8
+ # Define a function to process the output and extract only the transcription text
9
  def process_transcription(audio_input):
10
+ model = gr.Interface.load("models/openai/whisper-large-v3-turbo")
 
11
  result = model(audio_input)
12
 
13
+ # Extract the transcription text directly
14
+ transcription = result["text"]
15
+ return transcription
 
 
 
 
 
 
 
16
 
17
+ # Launch the interface
18
+ gr.Interface(
19
+ fn=process_transcription,
20
+ inputs="audio",
21
+ outputs="text",
22
+ live=True,
23
+ ).launch()