Speech Recognition with Amplitude and Frequency Modulations

Amplitude modulation (AM) and frequency modulation (FM) are commonly used in communication, but their relative contributions to speech recognition have not been fully explored. To bridge this gap, we derived slowly varying AM and FM from speech sounds and conducted listening tests using stimuli with...

Full description

Saved in:
Bibliographic Details
Published inProceedings of the National Academy of Sciences - PNAS Vol. 102; no. 7; pp. 2293 - 2298
Main Authors Zeng, Fan-Gang, Nie, Kaibao, Stickney, Ginger S., Kong, Ying-Yee, Vongphoe, Michael, Bhargave, Ashish, Wei, Chaogang, Cao, Keli, Merzenich, Michael M.
Format Journal Article
LanguageEnglish
Published United States National Academy of Sciences 15.02.2005
National Acad Sciences
Subjects
Audio equipment
Automatic speech recognition
Biological Sciences
Cochlear Implants
Comparative analysis
Deafness - psychology
Deafness - therapy
Engineering
Female
Frequency modulation
Humans
Line spectra
Listening
Male
Neurology
Perceptual Masking
Physical Sciences
Radio frequency
Signal noise
Spectral bands
Speech Acoustics
Speech Perception
Speech recognition
Supernova remnants
Theatrical masks
Voice recognition
Online AccessGet full text

Cover

More Information
Summary:Amplitude modulation (AM) and frequency modulation (FM) are commonly used in communication, but their relative contributions to speech recognition have not been fully explored. To bridge this gap, we derived slowly varying AM and FM from speech sounds and conducted listening tests using stimuli with different modulations in normal-hearing and cochlear-implant subjects. We found that although AM from a limited number of spectral bands may be sufficient for speech recognition in quiet, FM significantly enhances speech recognition in noise, as well as speaker and tone recognition. Additional speech reception threshold measures revealed that FM is particularly critical for speech recognition with a competing voice and is independent of spectral resolution and similarity. These results suggest that AM and FM provide independent yet complementary contributions to support robust speech recognition under realistic listening situations. Encoding FM may improve auditory scene analysis, cochlear-implant, and audiocoding performance.
Bibliography:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-2
content type line 23
ObjectType-Article-1
ObjectType-Feature-2
Author contributions: F.-G.Z. designed research; F.-G.Z., K.N., G.S.S., Y.-Y.K., M.V., A.B., C.W., and K.C. performed research; F.-G.Z., K.N., G.S.S., Y.-Y.K., M.V., C.W., and K.C. analyzed data; and F.-G.Z. wrote the paper.
To whom correspondence should be addressed. E-mail: fzeng@uci.edu.
This paper was submitted directly (Track II) to the PNAS office.
Abbreviations: AM, amplitude modulation; FM, frequency modulation; SRT, speech reception threshold; SNR, signal-to-noise ratio; IEEE, Institute of Electrical and Electronic Engineers; CUNY, City University of New York; HINT, Hearing in Noise Test.
Edited by Michael M. Merzenich, University of California, San Francisco, CA
ISSN:0027-8424
1091-6490
DOI:10.1073/pnas.0406460102