app.py

from transformers import pipeline

model_id = "Teapack1/model_KWS"  # update with your model id
pipe = pipeline("automatic-speech-recognition", model=model_id)

import gradio as gr

title = "Keyword Spotting Wav2Vec2"
description = "Gradio demo for finetuned Wav2Vec2 model on a custom dataset to perform keyword spotting task. Classes are scene 1, scene 2, scene 3, yes, no and stop."


def transcribe_speech(filepath):
    output = pipe(
        filepath,
        max_new_tokens=256,
        generate_kwargs={
            "task": "transcribe",
            "language": "sinhalese",
        },  # update with the language you've fine-tuned on
        chunk_length_s=30,
        batch_size=8,
    )
    return output["text"]

demo = gr.Blocks()

mic_transcribe = gr.Interface(
    fn=transcribe_speech,
    inputs=gr.Audio(sources="microphone", type="filepath"),
    outputs=gr.outputs.Textbox(),
)

file_transcribe = gr.Interface(
    fn=transcribe_speech,
    inputs=gr.Audio(sources="upload", type="filepath"),
    outputs=gr.outputs.Textbox(),
)

with demo:
    gr.TabbedInterface(
        [mic_transcribe, file_transcribe],
        ["Transcribe Microphone", "Transcribe Audio File"],
    )

demo.launch(debug=True)