app.py

import whisper
import gradio as gr

model = whisper.load_model("medium")


def transcribe(audio):

    # 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)
    #ps_prob = probs["ps"]
    #print(f"ستاسو د وینا د پښتو ژبې احتمال: {ps_prob}")
    #print(f"Detected language: {max(probs, key=probs.get)}")

    # decode the audio
    options = whisper.DecodingOptions(fp16=False, language="ps")
    result = whisper.decode(model, mel, options)
    return result.text


gr.Interface(
    title='د لومړي ځل لپاره د پښتو اتوماتیک وینا پیژندنه (Pashto ASR)',
    fn=transcribe,
    inputs=[
        gr.inputs.Audio(source="microphone", type="filepath")
    ],
    outputs=[
        "textbox"
    ],
    live=True).launch()