25 lines
768 B
Python
25 lines
768 B
Python
import torch
|
|
import librosa
|
|
import numpy as np
|
|
import soundfile as sf
|
|
from scipy.io import wavfile
|
|
from IPython.display import Audio
|
|
from transformers import Wav2Vec2ForCTC, Wav2Vec2Tokenizer
|
|
|
|
|
|
tokenizer = Wav2Vec2Tokenizer.from_pretrained("facebook/wav2vec2-base-960h")
|
|
model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
|
|
|
|
file_name = 'my-audio.wav'
|
|
|
|
data = wavfile.read(file_name)
|
|
framerate = data[0]
|
|
sounddata = data[1]
|
|
time = np.arange(0,len(sounddata))/framerate
|
|
input_audio, _ = librosa.load(file_name, sr=16000)
|
|
input_values = tokenizer(input_audio, return_tensors="pt").input_values
|
|
logits = model(input_values).logits
|
|
predicted_ids = torch.argmax(logits, dim=-1)
|
|
transcription = tokenizer.batch_decode(predicted_ids)[0]
|
|
print(transcription)
|