Nippon Telegraph and Telephone Corporation

The University of Tokyo

Ntt Corporation

Toudai Tlo, Ltd.

Yutaka Kamamoto

Takehiro Moriya

Noboru Harada

Ryosuke Sugiura

Yusuke Hiwasaki

Masahiro Fukui

Hirokazu Kameoka

Shigeki Sagayama

Takuya Nishimoto

Takahito Kawanishi

Junji Watanabe

Jun Terada

Naotaka Shibata

Shigeru Kuwano

Takashi Sato

Keisuke Hasegawa

Kenichi Noguchi

Jouji Nakayama

Yoshifumi Shiraki

Kouichi Furukado

Junichi Nakajima

There is provided a technology that improves, in a case where there is a sound signal obtained from a different code that is different from a code from which a decoded sound signal is obtained and that is derived from the same sound signal, the decoded sound signal by using the sound signal obtained from the different code. A signal (hereinafter, referred to as an upmixed common signal) obtained by upmixing a decoded sound common signal obtained by downmixing a decoded sound signal of each channel is subjected to signal purification using a signal (hereinafter, referred to as an upmixed monaural decoded sound signal) obtained by upmixing a monaural decoded sound signal to thereby generate a purified upmixed signal, and in each channel, the upmixed common signal is subtracted from the decoded sound signal and the purified upmixed signal is added thereto, to thereby generate a purified decoded sound signal.

Sound signal refinement method, sound signal decode method, apparatus thereof, program, and storage medium

A coding method and a decoding method are provided which can use in combination a predictive coding and decoding method which is a coding and decoding method that can accurately express coefficients which are convertible into linear prediction coefficients with a small code amount and a coding and decoding method that can obtain correctly, by decoding, coefficients which are convertible into linear prediction coefficients of the present frame if a linear prediction coefficient code of the present frame is correctly input to a decoding device. A coding device includes: a predictive coding unit that obtains a first code by coding a differential vector formed of differentials between a vector of coefficients which are convertible into linear prediction coefficients of more than one order of the present frame and a prediction vector containing at least a predicted vector from a past frame, and obtains a quantization differential vector corresponding to the first code; and a non-predictive coding unit that generates a second code by coding a correction vector which is formed of differentials between the vector of the coefficients which are convertible into the linear prediction coefficients of more than one order of the present frame and the quantization differential vector or formed of some of elements of the differentials.

Coding device, decoding device, and method and program thereof

A sound signal purification method includes a monaural decoded sound upmixing step of obtaining, an n-th channel upmixed monaural decoded sound signal {circumflex over ( )}Xthat is a signal obtained by upmixing a monaural decoded sound signal {circumflex over ( )}Xfor each channel by an upmixing process using the monaural decoded sound signal {circumflex over ( )}Xand inter-channel relationship information that is information indicating a relationship between channels of stereo, and an n-th channel signal purification step of obtaining a sequence based on a value {tilde over ( )}x(t) obtained by adding a value α×{circumflex over ( )}x(t) obtained by multiplying an n-th channel purification weight αby a sample value {circumflex over ( )}x(t) of the n-th channel upmixed monaural decoded sound signal {circumflex over ( )}Xand a value obtained by multiplying a value (1−α) obtained by subtracting the n-th channel purification weight αfrom 1 by a sample value {circumflex over ( )}x(t) of the n-th channel decoded sound signal {circumflex over ( )}X.

Sound signal purification using decoded monaural signals

An autocorrelation calculation unitcalculates an autocorrelation R(i) from an input signal. A prediction coefficient calculation unitperforms linear prediction analysis by using a modified autocorrelation R'(i) obtained by multiplying a coefficient w(i) by the autocorrelation R(i). It is assumed here, for each order i of some orders i at least, that the coefficient w(i) corresponding to the order i is in a monotonically increasing relationship with an increase in a value that is negatively correlated with a fundamental frequency of the input signal of the current frame or a past frame.

Linear prediction analysis device, method, program, and storage medium

A downmix unitobtains downmix signals which are signals obtained by mixing input sound signals of a left channel input and input sound signals of a right channel input. In a case where the left channel is preceding, the downmix signals are determined to be used as is in a left channel subtraction gain estimation unitand a left channel signal subtraction unit, and delayed downmix signals are determined to be used in a right channel subtraction gain estimation unitand a right channel signal subtraction unit. In a case where the right channel is preceding, the downmix signals are determined to be used as is in the right channel subtraction gain estimation unitand the right channel signal subtraction unit, and delayed downmix signals are determined to be used in the left channel subtraction gain estimation unitand the left channel signal subtraction unit

Sound signal encoding method, sound signal decoding method, sound signal encoding apparatus, sound signal decoding apparatus, program, and recording medium

For each frame, an n-th channel compensated decoded sound signalX'is obtained that is a signal obtained by compensating a high frequency of an n-th channel purified decoded sound signalXobtained by performing signal processing in a time domain on an n-th channel decoded sound signal {circumflex over ( )}Xobtained by decoding a stereo code CS. At this time, {circumflex over ( )}Xis selected in a case where the number of bits bcorresponding to an n-th channel in the number of bits of CS is larger than the number of bits bof a monaural code CM, or a monaural decoded sound signal obtained by decoding CM or a signal obtained by upmixing the monaural decoded sound signal for the n-th channel is selected in a case where bis smaller than b, an n-th channel high-frequency compensation gain ρthat is a value for bringing the high-frequency energy ofX′close to the high-frequency energy of {circumflex over ( )}Xis obtained, and a signal obtained by addingXand a signal obtained by multiplying a high-frequency component of the selected signal by ρis obtained and output as an n-th channel compensated decoded sound signal.

Sound signal high frequency compensation method, sound signal post processing method, sound signal decode method, apparatus thereof, program, and storage medium

For each frame, an n-th channel compensated decoded sound signal {tilde over ( )}X'is obtained that is a signal obtained by compensating a high frequency of an n-th channel purified decoded sound signal {tilde over ( )}Xobtained by performing signal processing in a time domain on an n-th channel decoded sound signal {circumflex over ( )}Xthat is a decoded sound signal of each channel of stereo obtained by decoding a stereo code CS. At this time, for the each frame with respect to the each channel, an n-th channel high-frequency compensation gain ρthat is a value for bringing high-frequency energy of {tilde over ( )}X′close to high-frequency energy of {circumflex over ( )}Xis obtained, and for the each frame with respect to the each channel, a signal obtained by adding {tilde over ( )}Xand a signal obtained by multiplying a high-frequency component of a monaural decoded sound signal that is obtained by decoding a monaural code CM that is a code different from the stereo code CS or a signal obtained by upmixing, for the each channel, the monaural decoded sound signal by the n-th channel high-frequency compensation gain ρis obtained and output as the n-th channel compensated decoded sound signal {tilde over ( )}X′.

Provided is a technique according to which it is possible to obtain a decoded sound signal of high sound quality without significantly increasing the delay time compared to a configuration in which only a decoded sound signal of the minimum necessary sound quality is obtained. In a terminal apparatus connected to a first communication line and a second communication line with a lower priority level there than, sound signals of multiple channels are obtained and output based on a monaural code included in a first code string input from the first communication line and an extended code included in a second code string with the closest frame number to that of the monaural code among extended codes included in the second code string input from the second communication line.

Sound signal receiving and decoding method, sound signal decoding method, sound signal receiving side apparatus, decoding apparatus, program and storage medium

A sound signal downmix method includes an inter-channel relationship information obtaining step of obtaining an inter-channel correlation value and preceding channel information of every pair of two channels included in N channels, the inter-channel correlation value being a value indicating a degree of a correlation between input sound signals of the two channels, the preceding channel information being information indicating which of the input sound signals of the two channels is preceding, and a downmix step of obtaining a downmix signal by weighting and adding the input sound signals of the N channels, the input sound signal of each channel being weighted based on the inter-channel correlation value and the preceding channel information such that the larger a correlation with an input sound signal of a preceding channel that precedes the channel, the smaller a weight, whereas the larger a correlation with an input sound signal of a succeeding channel that succeeds the channel, the larger the weight. The sum of the weights is normalized to 1.

Sound signal downmixing method, sound signal coding method, sound signal downmixing apparatus, sound signal coding apparatus, program and recording medium

For each frame, an n-th channel compensated decoded sound signalX'is obtained that is a signal obtained by compensating a high frequency of an n-th channel purified decoded sound signalXobtained by performing signal processing in a time domain on an n-th channel decoded sound signal {circumflex over ( )}X(n is each integer of 1 or more and N or less) that is a decoded sound signal of each channel of stereo obtained by decoding a stereo code CS. At this time, there are executed an n-th channel high-frequency compensation gain estimation step of obtaining an n-th channel high-frequency compensation gain ρthat is a value for bringing high-frequency energy of the n-th channel compensated decoded sound signalX′close to high-frequency energy of the n-th channel decoded sound signal {circumflex over ( )}X, and an n-th channel high-frequency compensation step of obtaining and outputting a signal obtained by adding the n-th channel purified decoded sound signalXand a signal obtained by multiplying a high-frequency component of the n-th channel decoded sound signal {circumflex over ( )}Xby the n-th channel high-frequency compensation gain ρ, as the n-th channel compensated decoded sound signalX′.

A downmix unitobtains downmix signals which are signals obtained by mixing input sound signals of a left channel input and input sound signals of a right channel input. A left channel signal subtraction unitand a right channel signal subtraction unitcode the difference between the input sound signals and a multiplication value of the downmix signals and a subtraction gain for each of the left channel and the right channel. In such a configuration, a left channel subtraction gain estimation unitand a right channel subtraction gain estimation unitdetermine the subtraction gain such that the quantization errors resulting from the two processes of coding/decoding are reduced.

A sound signal downmix method includes an inter-channel relationship information obtaining step of obtaining an inter-channel correlation value and preceding channel information of every pair of two channels included in N channels, the inter-channel correlation value being a value indicating a degree of a correlation between input sound signals of the two channels, the preceding channel information being information indicating which of the input sound signals of the two channels is preceding, and a downmix step of obtaining a downmix signal by weighting and adding the input sound signals of the N channels, the input sound signal of each channel being weighted based on the inter-channel correlation value and the preceding channel information such that the larger a correlation with an input sound signal of a preceding channel that precedes the channel, the smaller a weight, whereas the larger a correlation with an input sound signal of a succeeding channel that succeeds the channel, the larger the weight.

A sound signal downmix method includes an inter-channel relationship information obtaining step of obtaining an inter-channel correlation value and an inter-channel time difference in an approximate manner, and a downmix step of obtaining a downmix signal based on the obtained information. In the inter-channel relationship information obtaining step, multiple channel signals are sorted such that signals of adjacent channels are similar to each other, the inter-channel correlation value and the inter-channel time difference are determined only between adjacent channels after the sorting, the inter-channel correlation value between non-adjacent channels is obtained by a product or a geometric mean of the inter-channel correlations between the adjacent channels, and the inter-channel time difference between non-adjacent channels is obtained by adding up the inter-channel time differences of adjacent channels.

Growing community of inventors

Atsugi, Japan

Yutaka Kamamoto