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

transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-base.en")

def convert_to_mono(y, sr):
    # Check if the audio is already mono
    if len(y.shape) == 1:
        return y
    # Convert stereo to mono
    mono_y = librosa.to_mono(y.T)
    return mono_y

def transcribe(stream, new_chunk):
    sr, y = new_chunk
    y = y.astype(np.float32)
    y /= np.max(np.abs(y))
    
    # Convert audio to mono if it is stereo
    y = convert_to_mono(y, sr)

    if stream is not None:
        stream = np.concatenate([stream, y])
    else:
        stream = y
    return stream, transcriber({"sampling_rate": sr, "raw": stream})["text"]


demo = gr.Interface(
    transcribe,
    ["state", gr.Audio(sources=["microphone"], streaming=True)],
    ["state", "text"],
    live=True,
)

demo.launch()