Text-only Domain Adaptation using Unified Speech-Text Representation in Transducer

Open in new window