A prosodic Text-to-Speech system for Yorùbá language

Author

Akinwonmi, Akintoba Emmanuel ; Alese, B.K.

Author_Institution

Dept. of Comput. Sci., Fed. Univ. of Technol., Akure, Akure, Nigeria

fYear

2013

fDate

9-12 Dec. 2013

Firstpage

630

Lastpage

635

Abstract

We reviewed prosody realization in Text-to-Speech (TTS) applications. Standard Yoruba (SY) was considered alongside. Our review showed that the language is under-researched in the area of prosody and speech synthesis generally. Some technologies that produced good prosody in certain syllable based languages were identified. These include the use of polysyllabic units, diphones and Hidden Markov´s Model (HMM). We recommended a fusion of these technologies to realize the Yoruba TTS of high prosody. A mean opinion score (MOS) is also proposed to ascertain naturalness of the proposed TTS. This report presents among others, the review of prosody techniques in TTS, challenges of a Yoruba TTS and the proposed solution.

Keywords

hidden Markov models; natural language processing; speech synthesis; HMM; MOS; TTS applications; Yoruba TTS; Yoruba language; diphones; hidden Markov model; mean opinion score; polysyllabic units; prosodic text-to-speech system; prosody techniques; speech synthesis; syllable based languages; Databases; Estimation; Hidden Markov models; Speech; Syntactics; Prosody; Speech synthesis; Text-to-Speech; Yoruba TTS;

fLanguage

English

Publisher

ieee

Conference_Titel

Internet Technology and Secured Transactions (ICITST), 2013 8th International Conference for

Conference_Location

London

Type

conf

DOI

10.1109/ICITST.2013.6750279

Filename

6750279