Preprint / Version 2

Consonant Voicing Investigation

##article.authors##

  • Chen

DOI:

https://doi.org/10.31224/osf.io/w3tnr

Keywords:

CS.AI.SPEECH

Abstract

Obstruct consonant could be detected by energy difference profiles. In thisproject, we investigate algorithm to detect three types of landmarks, [stopclosure], [stop release], [fricative closure]. As a result, Overall detection ratesfor stop closure and release are 71.3% and 82.3% for obstruent landmarks inTIMIT, and fricatives yield 83.2% and 81.9% respectively.

Downloads

Download data is not yet available.

Posted

2019-08-09 — Updated on 2019-08-09

Versions