A modified frequency-weighted Itakura spectral distortion measure

Author

Li, Jian ; Krishnamurthy, Ashok K.

Author_Institution

Dept. of Electr. Eng., Ohio State Univ., Columbus, OH, USA

Volume

37

Issue

10

fYear

1989

fDate

10/1/1989 12:00:00 AM

Firstpage

1614

Lastpage

1617

Abstract

A modified frequency-weighted Itakura spectral distortion measure (FWIDM) for speech recognition in noise is described. It is shown that if noise is added to the test speech signal before preemphasis (such as would be the case with environment noise picked up at the microphone), the FWIDM method is not effective in reducing the distance between the noisy test spectrum and the clean reference spectrum. The modified FWIDM overcomes this problem. The modification is to use the bandwidth broadened spectrum of the nonpreemphasized test speech signal as the weighting function. It is shown that this weighting function reduces the distance between the clean reference spectrum and the noisy test spectrum of the preemphasized speech signals. This weighting function is also effective if the test and reference spectra are obtained from the nonpreemphasized speech signal

Keywords

encoding; speech recognition; encoding; frequency-weighted Itakura spectral distortion measure; noise; speech recognition; speech signal; weighting function; Bandwidth; Distortion measurement; Frequency measurement; Microphones; Noise measurement; Noise reduction; Speech enhancement; Speech recognition; Testing; Working environment noise;

fLanguage

English

Journal_Title

Acoustics, Speech and Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

0096-3518

Type

jour

DOI

10.1109/29.35406

Filename

35406