Anti-noise speech feature extraction and optimization based on HMM/SVM

LI Wan-ling; ZHANG Qiu-ju

doi:10.13873/J.1000-9787(2019)04-0055-04

Chinese

您当前的位置：

首页 >

文章列表页 >

Anti-noise speech feature extraction and optimization based on HMM/SVM

更新时间：2025-09-11

- Anti-noise speech feature extraction and optimization based on HMM/SVM
- Issue 4, Pages: 55-58(2019)
- 作者机构：
  
  1. 江南大学机械工程学院
  2. 江苏省食品先进制造装备技术重点实验室
- 作者简介：
- 基金信息：
- DOI：10.13873/J.1000-9787(2019)04-0055-04
  CLC： TN912.3;TP181
- Published：2019
- 稿件说明：
移动端阅览
LI Wan-ling, ZHANG Qiu-ju. Anti-noise speech feature extraction and optimization based on HMM/SVM[J]. 2019, (4): 55-58.
DOI：

LI Wan-ling, ZHANG Qiu-ju. Anti-noise speech feature extraction and optimization based on HMM/SVM[J]. 2019, (4): 55-58. DOI： 10.13873/J.1000-9787(2019)04-0055-04.

摘要

为了提高语音识别的鲁棒性

提出一种新的特征组合方法。方法基于F比对梅尔频率倒谱系数(MFCC)进行加权优化

同时将不同特征组合输入到语音隐马尔科夫模型(HMM)进行训练

得到具有抗噪性的最佳组合

并采用主成分分析(PCA)进行降维

增加支持向量机(SVM)分类器作为后处理器。实验表明

改进的MFCC、短时平均能量和Teager能量算子组合参数识别效果最优

识别率达到90. 48%。PCA降维后识别率降低了0. 4%

提升了计算速度。增加后处理器

系统识别率达到95. 25%

提高了系统的识别效率和分类决策力

相对于常规识别方法

准确率有所提高。

Abstract

In order to improve robustness of speech recognition

a new method of feature combination is proposed.Weighted optimization of Mel frequency cepstral coefficients（ MFCC） is carried out based on the F ratio.Meanwhile

several features are extracted for combination in different ways as input of hidden Markov model（ HMM） for training. Then

the best combination with anti-noise property is descended by PCA. Finally

the SVM classifier is added as a post processor. The experiment shows that the combination of improved MFCC

short-time average energy and Teager energy operator performs best

whose recognition rate is 90. 48 %. And dimensionality reduction based on PCA brings a 0. 4 % reduction on recognition rate. But the speed of calculation is promoted.After adding post processor

the total recognition rate of the system reaches 95. 25 %. The recognition efficiency of the system and the ability of system classification and decision-making are improved. Compared with the conventional recognition method

the accuracy rate is improved.

关键词

Keywords

references

Views

446

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Diagnosis algorithm for diastolic heart sound based on Hilbert marginal energy

Application of an improved feature extraction method in speech recognition

Study on fault diagnosis for spectroscope amplifier based on GSA-SVM

Research on sleeping posture recognition method based on flexible pressure sensor

Gearbox fault diagnosis based on multi-channel input CNN-SVM

Related Author

REN Jing

ZHAO Zhi-dong

CHEN Shu

YU Hai-bo

LIU Dongmei

WANG Haoran

LIU Chun

LIU Zhiqiang

Related Institution

School of Communication Engineering,Hangzhou Dianzi University

Hang Dian Smart City Research Center of Zhejiang Province

School of Internet of Things Engineering,Jiangnan University

School of Electrical and Automation Engineering, Hefei University of Technology

School of Mechanical Engineering,Hebei University of Technology

⁰