app.py

import gradio as gr
from transformers import pipeline

# Whisper model
asr = pipeline("automatic-speech-recognition", model="distil-whisper/distil-small.en")

# Summarization model
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")

def process_audio(audio_path):
    try:
        # Transcribe Audio
        transcription = asr(audio_path, return_timestamps=True)
        transcribed_text = transcription["text"]

        # Summarize Transcription
        summary = summarizer(transcribed_text, max_length=150, min_length=50, do_sample=False)[0]["summary_text"]

        return transcribed_text, summary
    except Exception as e:
        print(f"Error: {str(e)}")  # Log errors
        return f"Error: {str(e)}", ""

# Gradio Interface
iface = gr.Interface(
    fn=process_audio,
    inputs=gr.Audio(source="upload", type="filepath"),
    outputs=[gr.Textbox(label="Transcription"), gr.Textbox(label="Summary")],
    title="Audio Summarizer",
    description="Upload an audio file, and this app will transcribe and summarize its content.",
)

iface.launch()