Adapting Offline Speech Translation Models for Streaming with Future-Aware Distillation and Inference

Open in new window