@article{tokensplitusingdiscretespeech, title = {TokenSplit: Using Discrete Speech Representations for Direct, Refined, and Transcript-Conditioned Speech Separation and Recognition}, author = {Hakan Erdogan and Scott Wisdom and Xuankai Chang and Zalán Borsos and Marco Tagliasacchi and Neil Zeghidour and John R. Hershey}, year = {2023}, eprint = {2308.10415}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2308.10415v1}, }