Modelwav2vec2
wav2vec 2.0
Meta（Facebook AI Research）提出的自监督语音表征学习模型，通过对比学习在未标注音频数据上进行预训练，显著降低语音识别对标注数据的依赖
Timeline (last 90 days)
Sep 12
Meta's MMS model is based on the wav2vec 2.0 architecture and achieves speech recognition coverage by using religious texts (primarily multilingual translations of the Bible) as weakly supervised data sources
Unverified50%