legacy-datasets/common_voice
Updated • 1.37k • 144
How to use seccily/wav2vec-lt-lite with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("automatic-speech-recognition", model="seccily/wav2vec-lt-lite") # Load model directly
from transformers import AutoProcessor, AutoModelForCTC
processor = AutoProcessor.from_pretrained("seccily/wav2vec-lt-lite")
model = AutoModelForCTC.from_pretrained("seccily/wav2vec-lt-lite")YAML Metadata Error:"model-index[0].results[0].metrics[0].value" is required
The model can be used directly (without a language model) as follows:
import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
test_dataset = load_dataset("common_voice", "lt", split="test[:2%]")
processor = Wav2Vec2Processor.from_pretrained("seccily/wav2vec-lt-lite")
model = Wav2Vec2ForCTC.from_pretrained("seccily/wav2vec-lt-lite")
resampler = torchaudio.transforms.Resample(48_000, 16_000)
Test Result: 59.47