app.py

import os
os.system("pip install git+https://github.com/openai/whisper.git")
import gradio as gr
import whisper

def speech_to_text(tmp_filename, model_size):
    model = whisper.load_model(model_size)

    # load audio and pad/trim it to fit 30 seconds
    audio = whisper.load_audio(tmp_filename)
    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()
    result = whisper.decode(model, mel, options)

    # print the recognized text
    print(result.text)
    return result.text


gr.Interface(
    theme="dark",
    title="Whisper by OpenAI",
    thumbnail="https://cdn.openai.com/whisper/asr-summary-of-model-architecture-desktop.svg",
    css="""
    .gr-prose p{text-align: center;}
    .gr-button {background: black;color: white}
    """,
    description="Whisper is an automatic speech recognition (ASR) system trained on 680,000 hours of multilingual and multitask supervised data collected from the web.",
    fn=speech_to_text,
    inputs=[
        gr.Audio(label="Record your voice on your mic",source="microphone", type="filepath"),
        gr.Dropdown(label="Select model size",value="base",choices=["tiny", "base", "small", "medium", "large"])],
    outputs="text").launch()