Spaces:
Sleeping
Sleeping
File size: 1,991 Bytes
ee5ce46 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 |
from multilingual_translation import text_to_text_generation
from utils import lang_ids, data_scraping
import whisper
import gradio as gr
lang_list = list(lang_ids.keys())
model_list = data_scraping()
model = whisper.load_model("small")
def transcribe(audio):
#time.sleep(3)
# load audio and pad/trim it to fit 30 seconds
audio = whisper.load_audio(audio)
audio = whisper.pad_or_trim(audio)
# make log-Mel spectrogram and move to the same device as the model
mel = whisper.log_mel_spectrogram(audio).to(model.device)
# detect the spoken language
_, probs = model.detect_language(mel)
print(f"Detected language: {max(probs, key=probs.get)}")
# decode the audio
options = whisper.DecodingOptions(fp16 = False)
result = whisper.decode(model, mel, options)
finalResult = text_to_text_generation(prompt='return.text', model_id='facebook/m2m100_418M', device='cpu',target_lang='English')
return finalResult
# api endpoint to return the transcription in EN as a json response
# @app.route('/transcribe', methods=['POST'])
# def transcribe_api():
# if request.method == 'POST':
# audio = request.files['audio']
# audio = audio.read()
# audio = io.BytesIO(audio)
# audio = whisper.load_audio(audio)
# audio = whisper.pad_or_trim(audio)
# mel = whisper.log_mel_spectrogram(audio).to(model.device)
# _, probs = model.detect_language(mel)
# print(f"Detected language: {max(probs, key=probs.get)}")
# options = whisper.DecodingOptions(fp16 = False)
# result = whisper.decode(model, mel, options)
# return jsonify(result)
gr.Interface(
title = 'OpenAI Whisper ASR Gradio Web UI',
fn=transcribe,
inputs=[
gr.inputs.Audio(source="microphone", type="filepath")
],
outputs=[
"textbox"
],
live=True).launch(debug=True, enable_queue=True)
# output = gr.outputs.Textbox(label="Output Text") |