###
工程科学与技术:2014,46(2):127-132
←前一篇   |   后一篇→
本文二维码信息
码上扫一扫!
腭裂语音高鼻音等级自动识别算法研究
(1.四川大学 电气信息学院;2.四川大学 华西口腔医院)
Automatic Hypernasal Detection Based on Acoustic Analysis in Cleft Palate Speech
(1.School of Electrical Eng. and Info.,Sichuan Univ.;2.West China Hospital of Stomatology,Sichuan Univ.)
摘要
图/表
参考文献
相似文献
本文已被:浏览 1745次   下载 0
投稿时间:2013-06-14    修订日期:2014-01-01
中文摘要: 为了对腭裂语音的高鼻音进行等级区分,提出基于声学特征参数分析的腭裂语音高鼻音等级自动识别算法,提取基于香农能量和Mel倒谱系数(Mel frequency cepstrum coefficient,MFCC)的S-MFCC作为声学特征参数,结合高斯混合模型(Gaussian mixture model,GMM)分类器实现对腭裂语音4类高鼻音等级(正常、轻度、中度和重度)的自动识别。实验结果表明,提出的自动识别算法取得了较高的高鼻音类别正确识别率,对4类高鼻音的平均识别率达到79%以上,其中,提出的S-MFCC参数取得了85%的平均正确识别率,优于传统的香农能量算法、MFCC算法,具有较高的临床应用价值。
Abstract:In order to detect hypernasal automatically for cleft palate patients, based on Shannon energy and Mel frequency cepstrum coefficient acoustic features and by combining with Gaussian mixture model classifier, an automatic hypernasal detection algorithm was proposed. The experiment results showed that the presented method achieved a good performance on the detection of four levels of hypernasal, such as normal, low-level, moderate-level and high-level. The average classification accuracies for four levels of hypernasal were over 79%. Moreover, the correct recognition accuracy using energy plus Mel frequency cepstrum coefficient feature set reached up to 85%. The classification of hypernasal levels has important clinical applications.
文章编号:201300591     中图分类号:    文献标志码:
基金项目:国家自然科学基金青年基金资助项目(30900391)
作者简介:
引用文本:
何凌,袁亚南,尹恒,张桠童,张劲,刘奇,李杨.腭裂语音高鼻音等级自动识别算法研究[J].工程科学与技术,2014,46(2):127-132.
He Ling,Yuan Ya'nan,Yin Heng,Zhang Yatong,Zhang Jing,Liu Qi,Li Yang.Automatic Hypernasal Detection Based on Acoustic Analysis in Cleft Palate Speech[J].Advanced Engineering Sciences,2014,46(2):127-132.