File size: 2,144 Bytes
c106aba ea8b34f 38368a5 ea8b34f c106aba 38368a5 c106aba 06628a1 c106aba 38368a5 ea8b34f 38368a5 ea8b34f c106aba 06628a1 ea8b34f 06628a1 c106aba 38368a5 cb71106 06628a1 |
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 scipy.signal as sps
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)
def read_file(wav):
sample_rate, signal = wav_file
signal = signal.mean(-1)
number_of_samples = round(len(signal) * float(16000) / sample_rate)
resampled_signal = sps.resample(signal, number_of_samples)
return resampled_signal
def parse_transcription(wav_file):
'''filename = wav_file.name.split('.')[0]
convert(wav_file.name, filename + "16k.wav")
speech, _ = sf.read(filename + "16k.wav")
'''
speech = read_file(wav_file)
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
processor = Wav2Vec2Processor.from_pretrained("Harveenchadha/vakyansh-wav2vec2-hindi-him-4200")
model = Wav2Vec2ForCTC.from_pretrained("Harveenchadha/vakyansh-wav2vec2-hindi-him-4200")
processor = Wav2Vec2Processor.from_pretrained("Harveenchadha/vakyansh-wav2vec2-hindi-him-4200")
model = Wav2Vec2ForCTC.from_pretrained("Harveenchadha/vakyansh-wav2vec2-hindi-him-4200")
#input_ = gr.inputs.Audio(source="microphone", type="file")
input_ = gr.inputs.Audio(source="microphone", type="numpy")
gr.Interface(parse_transcription, inputs = input_, outputs="text",
analytics_enabled=False, show_tips=False, enable_queue=True).launch(inline=False); |