ResearchDatasetsTTSFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMultilingual Speech GenerationTTS multilingual (test)0.447WER60Spoken Instruction FollowingTTS Unseen (test)77LC Win Rate7Text-to-Speech GenerationTTS1.96WER4Audio GenerationTTS 80-ms audio segments (test)4.4Latency (ms)3Phonological faithfulness auditTTS16Mismatch Rate1