Fetching the paper…

Non-autoregressive End-to-end Speech Translation with Parallel Autoregressive Rescoring · Around