Abstract | ||
---|---|---|
The research focuses on the use of Hidden Markov Model (HMM) to build Khmer text-to-speech (TTS) system. Although the system is based on HMM statistic model, language specific functions were newly designed and developed to cope with the orthographical and grammatical nature of Khmer, some of which included word segmentation, grapheme to phoneme conversion, definitions of full context labels and question sets. In total four-thousand phonemically-balanced Khmer sentences were read aloud by an adult male speaker of Khmer, which were in turn served for training a model for Khmer TTS. The system has been incorporated into VoiceTra, a multilingual speech-to-speech translation app that has been developed and maintained by NICT. The app is publicly released for mobile devices and available to download in both App store and Google Play store. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/CIFEr.2019.8759128 | 2019 IEEE Conference on Computational Intelligence for Financial Engineering & Economics (CIFEr) |
Keywords | Field | DocType |
Khmer TTS,multilingual speech-to-speech translation app,HMM-based TTS system framework,Khmer text-to-speech system,HMM statistic model,language specific functions,word segmentation,phonemically-balanced Khmer sentences,hidden Markov model,VoiceTra,NICT,mobile devices,App store,Google Play store | System framework,Statistic,App store,Grapheme,Computer science,Text segmentation,Mobile device,Natural language processing,Artificial intelligence,Hidden Markov model | Conference |
ISSN | ISBN | Citations |
2380-8454 | 978-1-7281-0034-0 | 0 |
PageRank | References | Authors |
0.34 | 1 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Saly Keo | 1 | 0 | 0.34 |
Soky Kak | 2 | 0 | 0.34 |
Yoshinori Shiga | 3 | 45 | 13.35 |
Hiroaki Kato | 4 | 37 | 18.25 |
Hisashi Kawai | 5 | 250 | 54.04 |