Spaces:
Runtime error
Runtime error
File size: 2,397 Bytes
9fbe6a6 17f9ae1 9fbe6a6 c9292a7 9fbe6a6 17f9ae1 9fbe6a6 17f9ae1 c9292a7 9fbe6a6 e926ef7 17f9ae1 9fbe6a6 53e52ee b9f5cad ca3a196 5841b2a c9292a7 e926ef7 53e52ee |
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 |
import soundfile as sf
import torch
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import gradio as gr
import sox
import os
def convert(inputfile, outfile):
sox_tfm = sox.Transformer()
sox_tfm.set_output_format(
file_type="wav", channels=1, encoding="signed-integer", rate=16000, bits=16
)
sox_tfm.build(inputfile, outfile)
api_token = os.getenv("API_TOKEN")
model_name = "indonesian-nlp/wav2vec2-indonesian-javanese-sundanese"
processor = Wav2Vec2Processor.from_pretrained(model_name, use_auth_token=api_token)
model = Wav2Vec2ForCTC.from_pretrained(model_name, use_auth_token=api_token)
def parse_transcription(wav_file):
filename = wav_file.name.split('.')[0]
convert(wav_file.name, filename + "16k.wav")
speech, _ = sf.read(filename + "16k.wav")
input_values = processor(speech, sampling_rate=16_000, return_tensors="pt").input_values
logits = model(input_values).logits
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.decode(predicted_ids[0], skip_special_tokens=True)
return transcription
output = gr.outputs.Textbox(label="The transcript")
input_ = gr.inputs.Audio(source="microphone", type="file")
gr.Interface(parse_transcription, inputs=input_, outputs=[output],
analytics_enabled=False,
show_tips=False,
theme='huggingface',
layout='vertical',
title="Multilingual Speech Recognition for Indonesian Languages",
description="Automatic Speech Recognition Live Demo for Indonesian, Javanese and Sundanese Language",
article="Due to the issue with the current Gradio version, please use <a href='https://asr.ai-research.id/' target='_blank'>Multilingual Indonesian ASR</a> instead this one.<br>"
"This demo was built for the project "
"<a href='https://github.com/indonesian-nlp/multilingual-asr' target='_blank'>Multilingual Speech Recognition for Indonesian Languages</a>. "
"It uses the <a href='https://huggingface.co/indonesian-nlp/wav2vec2-indonesian-javanese-sundanese' target='_blank'>indonesian-nlp/wav2vec2-indonesian-javanese-sundanese</a> model "
"which was fine-tuned on Indonesian Common Voice, Javanese and Sundanese OpenSLR speech datasets.",
enable_queue=True).launch( inline=False)
|