app.py

from transformers import pipeline
import os
import gradio as gr
import torch
from IPython.display import Audio as IPythonAudio

#Audio to text
asr = pipeline(task="automatic-speech-recognition",
               model="distil-whisper/distil-small.en")
#Text to text
translator = pipeline(task="translation",
                      model="facebook/nllb-200-distilled-600M",
                      torch_dtype=torch.bfloat16) 
#Text to audio
pipe = pipeline("text-to-speech", model="suno/bark-small")

                
demo = gr.Blocks()
def transcribe_speech(filepath):
    if filepath is None:
        gr.Warning("No audio found, please retry.")
        return ""
    output = translator(asr(filepath)["text"],
                             src_lang="eng_Latn",
                             tgt_lang="hin_Deva")
    narrated_text=pipe(output[0]['translation_text'])
    output=IPythonAudio(narrated_text["audio"][0],
             rate=narrated_text["sampling_rate"])
    return output
    
mic_transcribe = gr.Interface(
    fn=transcribe_speech,
    inputs=gr.Audio(sources="microphone",
                    type="filepath"),
    outputs=gr.Audio(label="Translated Message"),
    allow_flagging="never")

file_transcribe = gr.Interface(
    fn=transcribe_speech,
    inputs=gr.Audio(sources="upload",
                    type="filepath"),
    outputs=gr.Audio(label="Translated Message"),
    allow_flagging="never",
)
with demo:
    gr.TabbedInterface(
        [mic_transcribe,
         file_transcribe],
        ["Transcribe Microphone",
         "Transcribe Audio File"],
    )

demo.launch(share=True)
demo.close()