app.py

import gradio as gr
from transformers import pipeline
import numpy as np

# Load the pre-trained models for transcription and summarization
asr_model = pipeline("automatic-speech-recognition", model="jonatasgrosman/wav2vec2-large-xlsr-53-english")
summarization_model = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6")

# Function to transcribe and summarize
def transcribe_and_summarize(audio):
    if audio is None:
        return "Error: No audio file provided.", None

    # Check if the audio is from a mic recording (tuple) or file (ndarray)
    if isinstance(audio, tuple):  # Mic recordings are returned as (sample_rate, data)
        audio = np.array(audio[1], dtype=np.float32)

    transcription = asr_model(audio)["text"]
    summary = summarization_model(transcription, max_length=130, min_length=30, do_sample=False)[0]["summary_text"]
    
    return transcription, summary

# Create a Gradio interface
interface = gr.Interface(
    fn=transcribe_and_summarize, 
    inputs=gr.Audio(type="filepath"), 
    outputs=["text", "text"],
    title="Meeting Transcription and Summarization",
    description="Upload an audio file or record using the mic to get a transcription and summary."
)

# Launch the interface
interface.launch()