专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

DE69609099D1 未知
公开(公告)号：DE69609099D1
公开(公告)日：2000-08-10
申请号：DE69609099
申请日：1996-03-07
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MORIYA TAKEHIRO , MANO KAZUNORI , MIKI SATOSHI , OHMURO HITOSHI , SASAKI SHIGEAKI , IWAKAMI NAOKI
IPC分类号： G10L19/00 , G10L19/04 , G10L19/06 , G10L19/12 , H03H17/02
摘要： In a CELP coding scheme, p-order LPC coefficients of an input signal are transformed into n-order LPC cepctrum coefficients cj (S2), which are modified into n-order modified LPC cepstrum coefficients cj' (S3). Log power spectral envelopes of the input signal and a masking function suited thereto are calculated (Figs. 3B, C), then they are subjected to inverse Fourier transform to obtain n-order LPC cepstrum coefficients, respectively, (Figs. 3D, E), then the relationship between each corresponding orders of the both LPC cepstrum coefficients is calculated, and the modification in step S3 is carried out on the basis of the relationship. The modified coefficients cj are inversely transformed by the method of least squares into m-order LPC coefficients for use as filter coefficients of a perceptual weighting filter. This concept is applicable to a postfilter as well.

2. 发明专利

DE69518452T2 未知
公开(公告)号：DE69518452T2
公开(公告)日：2001-04-12
申请号：DE69518452
申请日：1995-03-14
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： IWAKAMI NAOKI , MORIYA TAKEHIRO , MIKI SATOSHI
IPC分类号： G01R23/16 , G10L19/02 , G10L25/12 , G10L25/27 , G10L101/027 , G10L101/06 , G10L101/12
摘要： An input acoustic signal is subjected to modified discrete cosine transform processing to obtain its spectrum characteristics. On the other hand, linear prediction coefficients are derived from the input acoustic signal in a linear prediction coding analysis part (17), and the prediction coefficients are subjected to Fourier transform in a spectrum envelope calculation part (21) to obtain the envelope of the spectrum characteristics of the input acoustic signal. In a normalization part (22) the spectrum characteristics are normalized by the envelope thereof to obtain residual coefficients. A normalization part (26) normalizes the residual coefficients by a residual-coefficients envelope predicted in a residual-coefficients envelope calculation part (23), thereby obtaining fine structure coefficients, which are vector-quantized in a quantization part (25). A de-normalization part (31) de-normalizes the quantized fine structure coefficients. The residual-coefficients envelope calculation part (23) uses the reproduced residual coefficients to predict the envelope of residual coefficients of the subsequent frame.

3. 发明专利

DE69518452D1 未知
公开(公告)号：DE69518452D1
公开(公告)日：2000-09-28
申请号：DE69518452
申请日：1995-03-14
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： IWAKAMI NAOKI , MORIYA TAKEHIRO , MIKI SATOSHI
IPC分类号： G01R23/16 , G10L19/02 , G10L25/12 , G10L25/27 , G10L101/027 , G10L101/06 , G10L101/12
摘要： An input acoustic signal is subjected to modified discrete cosine transform processing to obtain its spectrum characteristics. On the other hand, linear prediction coefficients are derived from the input acoustic signal in a linear prediction coding analysis part (17), and the prediction coefficients are subjected to Fourier transform in a spectrum envelope calculation part (21) to obtain the envelope of the spectrum characteristics of the input acoustic signal. In a normalization part (22) the spectrum characteristics are normalized by the envelope thereof to obtain residual coefficients. A normalization part (26) normalizes the residual coefficients by a residual-coefficients envelope predicted in a residual-coefficients envelope calculation part (23), thereby obtaining fine structure coefficients, which are vector-quantized in a quantization part (25). A de-normalization part (31) de-normalizes the quantized fine structure coefficients. The residual-coefficients envelope calculation part (23) uses the reproduced residual coefficients to predict the envelope of residual coefficients of the subsequent frame.

4. 发明专利

DE69328450T2 未知
公开(公告)号：DE69328450T2
公开(公告)日：2001-01-18
申请号：DE69328450
申请日：1993-06-28
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MORIYA TAKEHIRO , KATAOKA AKITOSHI , MANO KAZUNORI , MIKI SATOSHI , OMURO HITOSHI , HAYASHI SHINJI
IPC分类号： G10L19/005 , G10L19/06 , G10L19/07 , G10L19/08 , G10L19/083 , G10L19/12 , G10L19/135 , G10L19/14
摘要： In a speech coding method of the present invention, initially, a plurality of samples of speech data are analyzed by a linear prediction analysis and thereby prediction coefficients are calculated. Then, the prediction coefficients are quantized, and the quantized prediction coefficients are set in a synthesis filter. Moreover, a pitch period vector is selected from an adaptive codebook in which a plurality of pitch period vector are stored, and the selected pitch period vector is multiplied by a first gain which is obtained, at the same time, with a second gain. In addition, a noise waveform vector is selected from a random codebook in which a plurality of the noise waveform vectors are stored, and is multiplied by a predicted gain and the second gain. Then, the speech vector is synthesized by exciting the synthesis filter with the pitch period vector multiplied by the first gain, and with the noise waveform vector multiplied by the predicted gain and the second gain. Consequently, speech data comprising a plurality of samples are coded as a unit of a frame operation. Furthermore, the predicted gain multiplied by the noise waveform vector which is selected in a subsequent frame operation, is predicted based on the current noise waveform vector which is multiplied by the predicted gain and the second gain at the current frame operation, and also the previous waveform vector which is multiplied by the predicted gain and the second gain in the previous frame operation.

5. 发明专利

DE69227401D1 未知
公开(公告)号：DE69227401D1
公开(公告)日：1998-12-03
申请号：DE69227401
申请日：1992-05-21
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MIKI SATOSHI , MORIYA TAKEHIRO , MANO KAZUNORI , OHMURO HITOSHI , SUDA HIROHITO
IPC分类号： G10L19/00 , G10L19/12 , G10L19/135 , G10L25/06 , G10L9/18 , G10L9/16 , G10L9/14
摘要： An excitation vector of the previous frame stored in an adaptive codebook is cut out with a selected pitch period. The excitation vector thus cut out is repeated until one frame is formed, by which a periodic component codevector is generated (S2). An optimum pitch period is searched for so that distortion of a reconstructed speech obtained by exciting a linear predictive synthesis filter with the periodic component codevector is minimized (S3, S4). Thereafter, a random codevector selected from a random codebook is cut out with the optimum pitch period and is repeated until one frame is formed, by which a repetitious random codevector is generated (S6). The random codebook is searched for a random codevector which minimizes the distortion of the reconstructed speech which is provided by exciting the synthesis filter with the repetitious random code vector (S7,S8).

6. 发明专利

DE69609099T2 未知
公开(公告)号：DE69609099T2
公开(公告)日：2001-03-22
申请号：DE69609099
申请日：1996-03-07
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MORIYA TAKEHIRO , MANO KAZUNORI , MIKI SATOSHI , OHMURO HITOSHI , SASAKI SHIGEAKI , IWAKAMI NAOKI
IPC分类号： G10L19/00 , G10L19/04 , G10L19/06 , G10L19/12 , H03H17/02
摘要： In a CELP coding scheme, p-order LPC coefficients of an input signal are transformed into n-order LPC cepctrum coefficients cj (S2), which are modified into n-order modified LPC cepstrum coefficients cj' (S3). Log power spectral envelopes of the input signal and a masking function suited thereto are calculated (Figs. 3B, C), then they are subjected to inverse Fourier transform to obtain n-order LPC cepstrum coefficients, respectively, (Figs. 3D, E), then the relationship between each corresponding orders of the both LPC cepstrum coefficients is calculated, and the modification in step S3 is carried out on the basis of the relationship. The modified coefficients cj are inversely transformed by the method of least squares into m-order LPC coefficients for use as filter coefficients of a perceptual weighting filter. This concept is applicable to a postfilter as well.

7. 发明专利

DE69328450D1 未知
公开(公告)号：DE69328450D1
公开(公告)日：2000-05-25
申请号：DE69328450
申请日：1993-06-28
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MORIYA TAKEHIRO , KATAOKA AKITOSHI , MANO KAZUNORI , MIKI SATOSHI , OMURO HITOSHI , HAYASHI SHINJI
IPC分类号： G10L19/005 , G10L19/06 , G10L19/07 , G10L19/08 , G10L19/083 , G10L19/12 , G10L19/135 , G10L19/14
摘要： In a speech coding method of the present invention, initially, a plurality of samples of speech data are analyzed by a linear prediction analysis and thereby prediction coefficients are calculated. Then, the prediction coefficients are quantized, and the quantized prediction coefficients are set in a synthesis filter. Moreover, a pitch period vector is selected from an adaptive codebook in which a plurality of pitch period vector are stored, and the selected pitch period vector is multiplied by a first gain which is obtained, at the same time, with a second gain. In addition, a noise waveform vector is selected from a random codebook in which a plurality of the noise waveform vectors are stored, and is multiplied by a predicted gain and the second gain. Then, the speech vector is synthesized by exciting the synthesis filter with the pitch period vector multiplied by the first gain, and with the noise waveform vector multiplied by the predicted gain and the second gain. Consequently, speech data comprising a plurality of samples are coded as a unit of a frame operation. Furthermore, the predicted gain multiplied by the noise waveform vector which is selected in a subsequent frame operation, is predicted based on the current noise waveform vector which is multiplied by the predicted gain and the second gain at the current frame operation, and also the previous waveform vector which is multiplied by the predicted gain and the second gain in the previous frame operation.

8. 发明专利

DE69227401T2 未知
公开(公告)号：DE69227401T2
公开(公告)日：1999-05-06
申请号：DE69227401
申请日：1992-05-21
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MIKI SATOSHI , MORIYA TAKEHIRO , MANO KAZUNORI , OHMURO HITOSHI , SUDA HIROHITO
IPC分类号： G10L19/00 , G10L19/12 , G10L19/135 , G10L25/06 , G10L9/18 , G10L9/16 , G10L9/14
摘要： An excitation vector of the previous frame stored in an adaptive codebook is cut out with a selected pitch period. The excitation vector thus cut out is repeated until one frame is formed, by which a periodic component codevector is generated (S2). An optimum pitch period is searched for so that distortion of a reconstructed speech obtained by exciting a linear predictive synthesis filter with the periodic component codevector is minimized (S3, S4). Thereafter, a random codevector selected from a random codebook is cut out with the optimum pitch period and is repeated until one frame is formed, by which a repetitious random codevector is generated (S6). The random codebook is searched for a random codevector which minimizes the distortion of the reconstructed speech which is provided by exciting the synthesis filter with the repetitious random code vector (S7,S8).

9. 发明专利

JPH1039900A ENVELOPE ESTIMATING METHOD OF FREQUENCY DOMAIN CONVERTING COEFFICIENT OF ACOUSTIC SIGNALS 失效
公开(公告)号：JPH1039900A
公开(公告)日：1998-02-13
申请号：JP19037596
申请日：1996-07-19
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： MORIYA TAKEHIRO , JIN AKIO , IWAGAMI NAOKI , MIKI SATOSHI , IKEDA KAZUNAGA
IPC分类号： G10L19/04 , G10L11/00 , H03M7/30 , H04B14/04 , G10L9/14 , G10L9/16
摘要： PROBLEM TO BE SOLVED: To precisely obtain the power envelope of MDCT coefficients. SOLUTION: The sample series of acoustic signals are superimposed for every N samples and taken out for every 2N samples (12). Then, these samples are multiplied by a Hanning function window (13). Samples x(i), (i=0,..., N/2-1) having an N/2 length of the beginning of a frame are reversed in time and substrated from next N/2 x(i), (i=N/2,..., N-1). Then, x(i), (i=N+N/2,..., 2N-1) having an N/2 length of the end of the frame are reversed in time, added to N/2 x(i), (i=N,..., N+N/2-1) just before, a folded process result y(i) is DCTed and MDCT coefficients are obtained (26). Moreover, y(i) is lenear prediction analyzed (24) and the envelopes of the MDCT coefficients are obtained from the power spectrum envelope of the predicted coefficients (28).

10. 发明专利

JPH09230898A ACOUSTIC SIGNAL TRANSFORMATION AND ENCODING AND DECODING METHOD 失效
公开(公告)号：JPH09230898A
公开(公告)日：1997-09-05
申请号：JP3520696
申请日：1996-02-22
申请人： NIPPON TELEGRAPH & TELEPHONE
发明人： IWAGAMI NAOKI , MORIYA TAKEHIRO , IKEDA KAZUNAGA , MIKI SATOSHI
IPC分类号： G10L19/08 , G10L19/00 , G10L19/02 , G10L19/04 , G10L19/12 , H03M7/30 , G10L9/18 , G10L9/14
摘要： PROBLEM TO BE SOLVED: To enable accurate quantization with a relatively smaller number of code vectors when transformation is performed in frequency domain for vector-quantization. SOLUTION: Frequency domain coefficients of an input acoustic signal are divided into small vectors, and corresponding weight coefficients are also divided similarly, and by using these corresponding ones, optimal code vectors are searched from a code book 39. In this case, the presence or absence of a minute structure is judged in the spectrum of the input acoustic signal, and if the minute structure exists, approximately a third of the absolute values among the element are left in descending order to the code vectors taken out of the code book 39 and are deformed so that the amplitudes become large by defining the other elements as zero, while, if the minute structure does not exist, they are used without performing the deformation.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式