File size: 3,048 Bytes
50547ae |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 |
import torch
import gradio as gr
import pytube as pt
from transformers import pipeline
from huggingface_hub import model_info
import time
import unicodedata
MODEL_NAME = "SakshiRathi77/wav2vec2-large-xlsr-300m-hi-kagglex"
lang = "hi"
device = 0 if torch.cuda.is_available() else "cpu"
pipe = pipeline(
task="automatic-speech-recognition",
model=MODEL_NAME,
device=device,
)
def transcribe(microphone, file_upload):
warn_output = ""
if (microphone is not None) and (file_upload is not None):
warn_output = (
"WARNING: You've uploaded an audio file and used the microphone. "
"The recorded file from the microphone will be used and the uploaded audio will be discarded.\n"
)
elif (microphone is None) and (file_upload is None):
return "ERROR: You have to either use the microphone or upload an audio file"
file = microphone if microphone is not None else file_upload
text = pipe(file)["text"]
return warn_output + text
def rt_transcribe(audio, state=""):
time.sleep(2)
text = pipe(audio)["text"]
state += unicodedata.normalize("NFC",text) + " "
return state, state
demo = gr.Blocks()
examples=[["examples/example1.mp3"], ["examples/example2.mp3"],["examples/example3.mp3"]]
title ="""
HindiSpeechPro: WAV2VEC-Powered ASR Interface
"""
description = """
<p>
<center>
Welcome to the HindiSpeechPro, a cutting-edge interface powered by a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice dataset.
<img src="https://huggingface.co/spaces/SakshiRathi77/SakshiRathi77-Wav2Vec2-hi-kagglex/resolve/main/Images/Hindi-Speech-Voice-Recognition-Tool.jpg" alt="logo" ;>
</center>
</p>
"""
# article = "<p style='text-align: center'><a href='https://github.com/SakshiRathi77/ASR' target='_blank'>Source Code on Github</a></p><p style='text-align: center'><a href='https://huggingface.co/blog/fine-tune-xlsr-wav2vec2' target='_blank'>Reference</a></p><p style='text-align: center'><a href='https://forms.gle/hjfc3F1P7m3weQVAA' target='_blank'><img src='https://e7.pngegg.com/pngimages/794/310/png-clipart-customer-review-feedback-user-service-others-miscellaneous-text-thumbnail.png' alt='Feedback Form' ;></a></p>"
mf_transcribe = gr.Interface(
fn=transcribe,
inputs=[
gr.inputs.Audio(source="microphone", type="filepath"),
gr.inputs.Audio(source="upload", type="filepath"),
],
outputs="text",
theme="huggingface",
title=title,
description= description ,
allow_flagging="never",
examples=examples,
)
rt_transcribe = gr.Interface(
fn=rt_transcribe,
inputs=[
gr.Audio(source="microphone", type="filepath", streaming=True),
"state"
],
outputs=[ "textbox",
"state"],
theme="huggingface",
title=title,
description= description ,
allow_flagging="never",
live=True,
)
with demo:
gr.TabbedInterface([mf_transcribe, rt_transcribe], ["Transcribe Audio", "Transcribe Realtime Voice"])
demo.launch(share=True)
|