Update app.py
Browse files
app.py
CHANGED
@@ -1,32 +1,24 @@
|
|
1 |
-
import subprocess
|
2 |
-
|
3 |
-
subprocess.run(["pip", "install", "datasets"])
|
4 |
-
subprocess.run(["pip", "install", "transformers"])
|
5 |
-
subprocess.run(["pip", "install", "torch", "torchvision", "torchaudio", "-f", "https://download.pytorch.org/whl/torch_stable.html"])
|
6 |
-
|
7 |
import gradio as gr
|
8 |
-
|
9 |
|
10 |
-
# Load model and processor
|
11 |
-
processor = WhisperProcessor.from_pretrained("openai/whisper-large")
|
12 |
-
model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large")
|
13 |
-
model.config.forced_decoder_ids = None
|
14 |
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
return
|
19 |
|
20 |
-
|
21 |
-
def
|
22 |
-
|
23 |
-
|
24 |
|
25 |
-
|
26 |
-
|
|
|
|
|
27 |
|
28 |
-
|
|
|
29 |
|
30 |
-
|
31 |
-
|
32 |
-
gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
+
import whisper
|
3 |
|
|
|
|
|
|
|
|
|
4 |
|
5 |
+
def transcribe_audio(audio_file):
|
6 |
+
model = whisper.load_model("base")
|
7 |
+
result = model.transcribe(audio_file)
|
8 |
+
return result["text"]
|
9 |
|
10 |
+
|
11 |
+
def main():
|
12 |
+
audio_input = gr.inputs.Audio(source="upload", type="filepath")
|
13 |
+
output_text = gr.outputs.Textbox()
|
14 |
|
15 |
+
iface = gr.Interface(fn=transcribe_audio, inputs=audio_input,
|
16 |
+
outputs=output_text, title="Audio Transcription App",
|
17 |
+
description="Upload an audio file and hit the 'Submit'\
|
18 |
+
button")
|
19 |
|
20 |
+
iface.launch()
|
21 |
+
|
22 |
|
23 |
+
if __name__ == '__main__':
|
24 |
+
main()
|
|