Voice assistants more and more use on-device Automated Speech Recognition (ASR) to make sure velocity and privateness. Nevertheless, attributable to useful resource constraints on the system, queries pertaining to advanced info domains typically require additional processing by a search engine. For such functions, we suggest a novel Transformer based mostly mannequin able to rescoring and rewriting, by exploring full context of the N-best hypotheses in parallel. We additionally suggest a brand new discriminative sequence coaching goal that may work properly for each rescore and rewrite duties. We present that our Rescore+Rewrite mannequin outperforms the Rescore-only baseline, and achieves as much as a mean 8.6% relative Phrase Error Price (WER) discount over the ASR system by itself.