基于改进语音特征与极限学习机的语音端点检测

罗庆; 包亚萍; 俞强

Chinese

您当前的位置：

首页 >

文章列表页 >

基于改进语音特征与极限学习机的语音端点检测

更新时间：2026-09-04

- 基于改进语音特征与极限学习机的语音端点检测
- Voice activity detection based on improved speech features and extreme learning machine
- 微电子学与计算机 2020年37卷第3期页码：37-41
- 作者机构：
  
  南京工业大学机构计算机科学与技术系,江苏,南京,211816
- 作者简介：
- 基金信息：
- DOI：
  中图分类号：
- 纸质出版：2020
- 稿件说明：
移动端阅览
罗庆, 包亚萍, 俞强. 基于改进语音特征与极限学习机的语音端点检测[J]. 微电子学与计算机, 2020, 37(3): 37-41.

LUO Qing, BAO Ya-ping, YU Qiang. Voice activity detection based on improved speech features and extreme learning machine[J]. 2020, 37(3): 37-41.
罗庆, 包亚萍, 俞强. 基于改进语音特征与极限学习机的语音端点检测[J]. 微电子学与计算机, 2020, 37(3): 37-41. DOI：

LUO Qing, BAO Ya-ping, YU Qiang. Voice activity detection based on improved speech features and extreme learning machine[J]. 2020, 37(3): 37-41. DOI：

摘要

语音端点检测(Voice Activity Detection

VAD)，是指在给定语音信号帧中判别语音是否存在，鲁棒的VAD有助于提高语音应用的自动化效率，例如语音增强、说话人识别、助听器等.为了提高低信噪比下语音端点检测的精度以及效率，提出了一种新的语音特征—低频消噪能量(Low Frequency De-noising Energy，LFDE)，将其应用于VAD中，并利用LFDE与现有的声学特征(梅尔频率倒谱参数、共振峰频率)结合训练极限学习机(Extreme Learning Machine

ELM)分类器.仿真实验发现，端点检测的精度与效率都有提高.

Abstract

Voice Activity Detection (VAD) refers to the determination of the existence of speech in a given speech signal frame. Robust VAD helps to improve the automation efficiency of speech applications

such as speech enhancement

speaker recognition

and hearing aids and so on. In order to improve the accuracy and efficiency of voice activity detection under low SNR

a new speech feature—Low Frequency De-noising Energy (LFDE) is proposed

which is applied to VAD and utilizes LFDE and existing acoustic features (Mel frequency cepstrum parameters

formants Frequency) combined with the Extreme Learning Machine (ELM) classifier.Simulation experiments show that the accuracy and efficiency of voice activity detection are improved.

关键词

Keywords

references

ZHAO H, GUO X, WANG M, et al. Analyze EEG signals with extreme learning machine based on PMIS feature selection [J]. International Journal of Machine Learning and Cybernetics, 2018, 9(2): 243-249. DOI: 243

SHIN J W, CHANG J H, KIM N S. Voice activity detection based on statistical models and machine learning approaches [J]. Computer Speech and Language, 2010, 24(3):515-530. DOI: 515

KORNIIENKO O, MACHUSKY E. Voice activity detection algorithm using spectral-correlation and wavelet-packet transformation [J]. Radioelectronics and Communications Systems, 2018, 61(5):185-193. DOI: 185

JAVIER RAMiREZ, PABLO YéLAMOS, JUAN MANUEL GǒRRIZ, et al. ELM-enabled Voice Activity Detection[C]// Advances in Neural Networks - ISNN 2006, Third International Symposium on Neural Networks. China, Chengdu, 2006.

PANG J. Spectrum energy based voice activity detection[C]// IEEE 2017 IEEE 7th Annual Computing and Communication Workshop and Conference (CCWC)-Las Vegas, NV, USA, 2017.

朱春利, 李昕.基于多特征融合与动态阈值的语音端点检测方法[J].计算机工程, 2019, 45(2):250-257. DOI: 250

MUKHERJEE H, OBAIDULLAH S M, PHADIKAR S, et al. MISNA - A musical instrument segregation system from noisy audio with LPCC-S features and extreme learning[J]. Multimedia Tools and Applications, 2018, DOI:10.1007/s11042-018-5993-6.

SHAHSAVARI S, SAMETI H, HADIAN H. Speech activity detection using deep neural networks[C]// Electrical Engineering. Tehran, Iran, IEEE, 2017.

沈希忠, 郑晓修.基于Teager能量算子和经验模态分解的语音端点检测算法[J].电子与信息学报, 2018, 40(7):93-99. DOI: 93

王宏志, 徐玉超, 李美静.基于Mel频率倒谱参数相似度的语音端点检测算法[J].吉林大学学报(工学版), 2012, 42(5):1331-1336. DOI: 1331

HUANG G B, ZHU Q Y, SIEW C K. Extreme learning machine: Theory and applications [J]. Neurocomputing, 2006, 70(1/3):489-501. DOI: 489

刘红星, 戴蓓蒨, 陆伟.基于共振峰谐波能量的语音端点检测[J].清华大学学报(自然科学版), 2008(s1):144-149. DOI: 144

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种基于特征上下文理解与切片分析的肺结核检测算法

基于MSS-YOLO-MPHands融合架构的装配手部动作识别研究

超低功耗实时时钟电路的设计

基于多尺度融合与小目标增强的航拍车辆检测算法

基于反向体偏置的高增益稳定性高温运放设计