app.py

from transformers import pipeline
import gradio as gr
import time

# Load the pipeline
pipe = pipeline("automatic-speech-recognition")  # You can specify model if needed

# Define the transcribe function
def transcribe(audio, state):
    time.sleep(1)  # simulate delay
    text = pipe(audio)["text"]
    state += text + " "
    return state, state

# Create the Gradio interface
with gr.Blocks() as demo:
    state = gr.State("")  # Proper state handling
    audio_input = gr.Audio(sources="microphone", type="filepath", label="Speak")
    output_text = gr.Textbox(label="Transcription")

    transcribe_button = gr.Button("Transcribe")

    transcribe_button.click(
        fn=transcribe,
        inputs=[audio_input, state],
        outputs=[output_text, state]
    )

demo.launch()