app.py

import gradio as gr
from transformers import pipeline
import os
os.system('git clone https://github.com/irshadbhat/indic-trans.git')
os.system('pip install ./indic-trans/.')
p1= pipeline(task="automatic-speech-recognition", model="cdactvm/w2v-bert-2.0-odia_v1")

def transcribe_odiya(speech):
    #print (p1(speech))
    text = p1(speech)["text"]
    #text=cleanhtml(text)
    return text
def transcribe_odiya_eng(speech):
    from indictrans import Transliterator
    trn = Transliterator(source='ori', target='eng', build_lookup=True)
    text = p1(speech)["text"]
    text=trn.transform(text)
    return text
    
def sel_lng(lng,mic=None, file=None):
    if mic is not None:
        audio = mic
    elif file is not None:
        audio = file
    else:
        return "You must either provide a mic recording or a file"
    if (lng=="Odiya"):
        return transcribe_odiya(audio)
    elif (lng=="Odiya-trans"):
        return transcribe_odiya_eng(audio)
    
    
        
demo=gr.Interface(
    fn=sel_lng, 
      
    inputs=[
        
        gr.Dropdown(["Odiya","Odiya-trans"],value="Odiya",label="Select Language"),
        gr.Audio(sources="microphone", type="filepath"),
        gr.Audio(sources="upload", type="filepath"),
        #"state"
    ],
    outputs=[
        "textbox"
#        #"state"
    ],
    title="Automatic Speech Recognition",
    description = "Demo for Automatic Speech Recognition. Use microphone to record speech. Please press Record button. Initially it will take some time to load the model. The recognized text will appear in the output textbox",
      ).launch()