app.py

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

transcriber_gujarati = pipeline("automatic-speech-recognition", model="ai4bharat/indicwav2vec_v1_gujarati")

def transcribe(audio):
    sr,y = audio
    y = y.astype(np.float32)
    y/= np.max(np.abs(y))
    return transcriber_gujarati(librosa.resample(y=y, orig_sr=sr, target_sr=16000))["text"]

gr.Interface(transcribe,inputs="microphone",outputs="text").launch()