NDLI: Improvement of speech signal extraction method using detection filter of energy spectrum entropy

Content Provider	Springer Nature Link
Author	Chung, Kyungyong Oh, SangYeob
Copyright Year	2015
Abstract	In speech recognition system research, recognition system performance has been significantly improved through research and development in the speech recognition area, but environmental noise is still a favorite subject for research due to its numerous environmental changes. And speech extraction techniques, which are widely applied, improve speech signals that are mixed with noise. A least mean square (LMS) adaptation filter is commonly used to help noise estimation and detection algorithms adapt to changing environments. But an LMS filter needs some time to adapt and estimate signals. That weakness can be overcome by using energy spectrum entropy and an average estimate LMS (AELMS) filter to detect robust voice activity in a noisy environment. In this paper, we propose a speech signal extraction method using a detection filter of energy spectrum entropy. The proposed method is polluted speech–signal noise extraction to reduce noise with an AELMS filter to detect robust voice activity. An AELMS filter maintains source features of speech, decreases speech information degradation, and reduces noise in a polluted speech signal. To improve adaptation speed, we calculated an average estimator, and controlled the LMS filter step size with a frame measure. For speech detection of signals synthesized with low-speed and high-speed driving noise, an energy spectrum entropy method was used. Compared to an existing method of using frame energy, the proposed method improved the starting point of the resulting speech by 1.7 % of an error rate and by 3.7 % of an end point error rate.
Starting Page	629
Ending Page	635
Page Count	7
File Format	PDF
ISSN	13867857
Journal	Cluster Computing
Volume Number	18
Issue Number	2
e-ISSN	15737543
Language	English
Publisher	Springer US
Publisher Date	2015-02-06
Publisher Place	New York
Access Restriction	One Nation One Subscription (ONOS)
Subject Keyword	Energy entropy Spectrum Noise estimation Frame energy LMS AELMS Processor Architectures Operating Systems Computer Communication Networks
Content Type	Text
Resource Type	Article
Subject	Computer Networks and Communications Software

Sl.	Authority	Responsibilities	Communication Details
1	Ministry of Education (GoI), Department of Higher Education	Sanctioning Authority	https://www.education.gov.in/ict-initiatives
2	Indian Institute of Technology Kharagpur	Host Institute of the Project: The host institute of the project is responsible for providing infrastructure support and hosting the project	https://www.iitkgp.ac.in
3	National Digital Library of India Office, Indian Institute of Technology Kharagpur	The administrative and infrastructural headquarters of the project	Dr. B. Sutradhar bsutra@ndl.gov.in
4	Project PI / Joint PI	Principal Investigator and Joint Principal Investigators of the project	Dr. B. Sutradhar bsutra@ndl.gov.in Prof. Saswat Chakrabarti will be added soon
5	Website/Portal (Helpdesk)	Queries regarding NDLI and its services	support@ndl.gov.in
6	Contents and Copyright Issues	Queries related to content curation and copyright issues	content@ndl.gov.in
7	National Digital Library of India Club (NDLI Club)	Queries related to NDLI Club formation, support, user awareness program, seminar/symposium, collaboration, social media, promotion, and outreach	clubsupport@ndl.gov.in
8	Digital Preservation Centre (DPC)	Assistance with digitizing and archiving copyright-free printed books	dpc@ndl.gov.in
9	IDR Setup or Support	Queries related to establishment and support of Institutional Digital Repository (IDR) and IDR workshops	idr@ndl.gov.in

Target speech feature extraction using non-parametric correlation coefficient

Improvement of Speech Detection Using ERB Feature Extraction

Acquiring lexical knowledge using raw corpora and unsupervised clustering method

An artifact detection scheme with CUDA-based image operations

Erratum to: Mobile image sensors for object detection using color segmentation

Adaptive Multiuser Channel Estimation using Reduced Kalman/LMS Algorithm

Object Placement Using Performance Surfaces

Special issue on energy-aware computing and communications

Serving a video into an image carousel: system design and implementation

Improvement of speech signal extraction method using detection filter of energy spectrum entropy

Similar Documents

Target speech feature extraction using non-parametric correlation coefficient

Improvement of Speech Detection Using ERB Feature Extraction

Acquiring lexical knowledge using raw corpora and unsupervised clustering method

An artifact detection scheme with CUDA-based image operations

Erratum to: Mobile image sensors for object detection using color segmentation

Adaptive Multiuser Channel Estimation using Reduced Kalman/LMS Algorithm

Object Placement Using Performance Surfaces

Special issue on energy-aware computing and communications

Serving a video into an image carousel: system design and implementation

Improvement of speech signal extraction method using detection filter of energy spectrum entropy