Line spectral frequency representation of subbands for speech recognition

Files
Line spectral frequency representation of subbands for speech recognition.pdf (247.9 KB)
Date
1995
Authors
Erzin, E.
Çetin, A.E.
Editor(s)
Advisor
Supervisor
Co-Advisor
Co-Supervisor
Instructor
Source Title
Signal Processing
Print ISSN
1651684
Electronic ISSN
Publisher
Volume
44
Issue
1
Pages
117 - 119
Language
English
Type
Article
Journal Title
Journal ISSN
Volume Title
Series
Abstract

In this paper, a new set of speech feature parameters is constructed from subband analysis based Line Spectral Frequencies (LSFs). The speech signal is divided into several subbands and the resulting subsignals are represented by LSFs. The performance of the new speech feature parameters, SUBLSFs, is compared with the widely used Mel Scale Cepstral Coefficients (MELCEPs). SUBLSFs are observed to be more robust than the MELCEPs in the presence of car noise. © 1995.

Course
Other identifiers
Book Title
Keywords
Line spectral frequency , Speech recognition , Feature extraction , Parameter estimation , Robustness (control systems) , Signal processing , Spectrum analysis , Spurious signal noise , Line spectral frequencies (LSF) , Subband analysis , Speech recognition
Citation
Permalink
http://hdl.handle.net/11693/25907
Published Version (Please cite this version)
http://dx.doi.org/10.1016/0165-1684(95)00038-F
Collections
Department of Electrical and Electronics Engineering