app.py

import whisper
import streamlit as st

st.title("Audio Transcription with Whisper")

# File uploader for audio
uploaded_file = st.file_uploader("Upload an audio file", type=["wav", "mp3", "m4a"])

if uploaded_file is not None:
    with open("output.mp3", "wb") as f:
        f.write(uploaded_file.getbuffer())
    st.audio("output.mp3")

    # Transcribe audio
    model = whisper.load_model("base")
    audio = whisper.load_audio("output.mp3")
    audio = whisper.pad_or_trim(audio)
    mel = whisper.log_mel_spectrogram(audio).to(model.device)

    # Detect the spoken language
    _, probs = model.detect_language(mel)
    st.write(f"Detected language: {max(probs, key=probs.get)}")

    # Decode the audio
    options = whisper.DecodingOptions()
    result = whisper.decode(model, mel, options)

    # Display the recognized text
    st.write(result.text)