2026-04-30T08:46:08Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/4678 2021-12-13T03:09:01Z 1582963302567:1597824273898 user-ucsy

Word-based Myanmar Text-to-Speech with CLUSTERGEN Hlaing, Aye Mya Pa, Win Pa In Myanmar language, word information isimportant for the naturalness of Myanmar speechbecause pronunciation changes depend on wordinformation. In this paper, Myanmar speechsynthesis with word information has beenimplemented by using CLUSTERGEN. Thecomparison of the naturalness of Myanmarsynthesized speech by using HMM-based model withsyllable information and CLUSTERGEN model withword information has been conducted based onComparison Mean Opinion Score (CMOS). Theresults show that word information can promote thenaturalness of the synthesized speech for Myanmarlanguage. 2018-02-22 http://hdl.handle.net/20.500.12678/0000004678 https://meral.edu.mm/records/4678