Kadazandusun speech recognition: a case study
Currently, there is no existing system that provides common information and utilities for Kadazandusun’s speech recognition since Kadazandusun speech has different features that are not available in other languages. This paper presents a preliminary experiment using one of the famous feature extract...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | en en |
| Published: |
UniSE Press
2021
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/33171/1/Kadazandusun%20speech%20recognition_%20a%20case%20study.pdf https://eprints.ums.edu.my/id/eprint/33171/3/Kadazandusun%20speech%20recognition_%20a%20case%20study%20_ABSTRACT.pdf https://eprints.ums.edu.my/id/eprint/33171/ http://tost.unise.org/pdfs/vol8/no3-3/ToST-CoFA2020-560-567-OA.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Currently, there is no existing system that provides common information and utilities for Kadazandusun’s speech recognition since Kadazandusun speech has different features that are not available in other languages. This paper presents a preliminary experiment using one of the famous feature extraction methods which is Linear Prediction Cepstral Coefficients (LPCC). Further investigation on the speech data is using several classifier algorithms to investigate the recognition rate of Kadazandusun words. There are 6 words of Kadazandusun collected as an individual speech to test the feature extraction and the classifiers. The objectives of this study are to investigate LPCC feature extraction and to propose a suitable classifier algorithm for Kadazandusun speech data. |
|---|