/Self-Supervised-Speech-Pretraining-and-Representation-Learning

Official implementation of the S3PRL toolkit: self-supervised pre-training of Mockingjay, TERA, AALBERT, APC, and more to come. With easy-to-use standard downstream evaluation scripts including phone classification, speaker recognition, and ASR. (All in Pytorch)

Primary LanguagePythonMIT LicenseMIT

Watchers