CA2513842A1

CA2513842A1 - Apparatus and method for speech coding

Info

Publication number: CA2513842A1
Application number: CA002513842A
Authority: CA
Inventors: Kazutoshi Yasunaga; Toshiyuki Morii
Original assignee: Individual
Current assignee: III Holdings 12 LLC
Priority date: 1999-08-23
Filing date: 2000-08-23
Publication date: 2001-03-01
Anticipated expiration: 2020-08-23
Also published as: CA2514249C; CA2514249A1; CA2513842C

Abstract

A speech encoder comprising an LPC synthesizing means, a gain calculating means and a parameter coding means for performing predictive coding of gains of an adaptive excitation vector and a stochastic excitation vector associated with indexes searched in the gain calculating means is provided. The parameter coding means comprises prediction coefficient adjusting means for adjusting one or more prediction coefficients used in the predictive coding according to one or more past subframe states. The speech encoder automatically adjusts prediction coefficients when the state value in a preceding subframe is an extremely large value or an extremely small value in predictive quantization with less including local abnormal sounds.

Claims

1. A speech encoder comprising:

LPC synthesizing means for obtaining a synthesized speech by filtering adaptive excitation vector and stochastic excitation vector stored in an adaptive codebook and stochastic codebook using an LPC coefficients obtained from an input speech;
gain calculating means for calculating gains of said adaptive excitation vector and said stochastic excitation vector and searching code of the adaptive excitation vector and searching code of the stochastic excitation vector using coding distortion between said input speech and said synthesized speech obtained using said gains; and parameter coding means for performing predictive coding of gains using the adaptive excitation vector and stochastic excitation vector corresponding to the codes obtained, wherein said parameter coding means comprises prediction coefficient adjusting means for adjusting one or more prediction coefficients used for said predictive coding according to one or more states of a previous subframe.

2. The speech encoder according to claim 1, wherein when one or more states of a previous subframe are an extremely large value or an extremely small value, said prediction coefficient adjusting means adjusts said prediction coefficients so as to reduce the influence thereof.

3. The speech encoder according to claim 1, wherein said parameter coding means has a codebook including gain vectors of the adaptive excitation vectors, logarithmic gain vectors of the stochastic excitation vectors and coefficients for adjusting the prediction coefficient.

4. The speech encoder according to claim 3, wherein in predicting coding when a product sum between states and prediction coefficients are calculated, prediction coefficient adjustment coefficients corresponding to the states are multiplied.

5. The speech encoder according to claim 1, further comprising storing means for storing said adaptive excitation vector, said stochastic excitation vector and prediction coefficient adjustment coefficient in accordance with each state.

6. The speech encoder according to claim 5, wherein when said adaptive excitation vector and said stochastic excitation vector stored in said storing means are updated, said prediction coefficient adjustment coefficients are also updated.

7. A CELP-based speech encoder that performs encoding by decomposing one frame into a plurality of subframes, comprising:

LPC synthesizing means for obtaining a synthesized speech by filtering adaptive excitation vector and stochastic excitation vector stored in an adaptive codebook and stochastic codebook using LPC coefficients obtained from an input speech;

gain calculating means for calculating gains of said adaptive excitation vector and said stochastic excitation vector; and parameter coding means for performing vector quantization of the adaptive excitation vector and stochastic excitation vector obtained using coding distortion between said input speech and said synthesized speech and said gains, and further comprising:

pitch analyzing means for performing pitch analyses of a plurality of subframes in the frame respectively, before performing an adaptive codebook search for the first subframe, finding a correlation value and calculating a value most approximate to the pitch period using said correlation values.

8. The speech encoder according to claim 7, further comprising search range setting means for determining a lag search range of a plurality of subframes based on the said adaptive codebook and said stochastic codebook using decoded LPC coefficients obtained from an input speech;
calculating gains of said adaptive excitation vector and said stochastic excitation vector;
performing vector quantization on the adaptive excitation vector and stochastic excitation vector determined using coding distortion between said input speech and said synthesized speech; and calculating correlation values by performing pitch analyses of a plurality of subframes in the processing frame before performing an adaptive codebook search of the first subframe and calculating a value most approximate to the pitch period using said correlation values.