Encoding of lexical tone in self-supervised models of spoken language

Open in new window