公开(公告)号：GB2357231A

公开(公告)日：2001-06-13

申请号：GB0023864

申请日：2000-09-29

Applicant: IBM

Inventor： HOORY RON ,  CHAZAN DAN ,  SILVERA EZRA ,  ZILBULSKI MEIR

IPC: G10L15/02 ,  G10L19/02

Abstract: In a method for encoding a digitized speech signal so as to generate data capable of being decoded as speech, a digitized speech signal is first converted to a series of feature vectors by deriving at successive instances of time, e.g. using ABS and Mel-Binning unit 32, an estimate of the spectral envelope of the digitized speech signal and multiplying each estimate of the spectral envelope by a predetermined set of frequency domain window functions, wherein each window occupies a narrow range of frequencies. The integrals thereof are computed and they or a set of predetermined functions thereof are assigned to respective components of a corresponding feature vector in the series of feature vectors. For each instance of time a respective pitch value of the digitized speech signal is computed at 34,35, and successive acoustic vectors each containing the respective pitch value and feature vector are compressed so as to derive therefrom a bit stream. A suitable decoder reverses the operation so as to extract the features vectors and pitch values, thus allowing speech reproduction and playback. In addition, speech recognition is possible using the decompressed feature vectors, with no impairment of the recognition accuracy and no computational overhead.

公开(公告)号：GB2357231B

公开(公告)日：2004-06-09

申请号：GB0023864

申请日：2000-09-29

Applicant: IBM

Inventor： HOORY RON ,  CHAZAN DAN ,  SILVERA EZRA ,  ZILBULSKI MEIR

IPC: G10L15/02 ,  G10L19/02

Abstract: A method for encoding a digitized speech signal so as to generate data capable of being decoded as speech. A digitized speech signal is first converted to a series of feature vectors using for example known Mel-frequency Cepstral coefficients (MFCC) techniques. At successive instances instance of time a respective pitch value of the digitized speech signal is computed, and successive acoustic vectors each containing the respective pitch value and feature vector are compressed so as to derive therefrom a bit stream. A suitable decoder reverses the operation so as to extract the features vectors and pitch values, thus allowing speech reproduction and playback. In addition, speech recognition is possible using the decompressed feature vectors, with no impairment of the recognition accuracy and no computational overhead.