Facebook claims wav2vec 2.0 tops speech recognition performance with 10 minutes of labeled data
In a paper published on the preprint server Arxiv.org, researchers at Facebook describe wav2vec 2.0, an improved framework for self-supervised speech recognition. They claim it demonstrates for the first time that learning representations from speech, followed by fine-tuning on transcribed speech, can outperform the best semi-supervised methods while being conceptually simpler, achieving state-of-the-art results using […]
