Université De Sherbrooke

International Business Machines Corporation

Claude Laflamme

Jean-Pierre Adoul

Tero Honkanen

Hua Ye

Sasan Ardalan

Ali Soheil Sadri

Redwan Salami

Redman Salami

A method and device for analog simultaneous voice and data transmission over an analog channel encodes at a voice signal sampling rate a sampled voice signal including spectral and pitch information. During encoding, the spectral and pitch information is extracted from the sampled voice signal to produce a residual signal, and the residual signal is converted into a sequence of residual vectors at a Baud rate. Then, a bitstream including non voice data and the extracted spectral and pitch information is produced. A modem unit, operating at Baud rate and producing an output carrier wave supplied to the analog channel, produces a sequence of symbol vectors in response to the sequence of residual vectors and the bitstream, and modulates the output carrier wave in response to the sequence of symbol vectors. A fractional sampling-rate conversion is conducted on the residual signal by shifting the voice signal sampling rate by a factor M/N, where M and N are the smallest integers that verify, or closely verify, the equality M/N=2R/F, where R is the Baud rate and F is the voice signal sampling rate. This fractional sampling-rate conversion further involves anti-aliasing filtering of the residual signal. In this manner, encoding operates at a voice signal sampling rate independent of the modem's Baud rate.

Baud-rate-independent ASVD transmission built around G.729 speech-coding

Instability inherent in analysis-by-synthesis speech/audio codecs and caused in particular by channel errors during transmission of highly periodic signals such as high-frequency sine waves is removed. Analysis-by-synthesis techniques involve production, in response to the speech/audio signal and at regular time intervals called frames, of (a) a set of spectral parameters for use in driving a synthesis filter in view of synthesizing the speech/audio signal, and (b) a pitch gain for constructing a past-excitation-signal component supplied to the synthesis filter. In accordance with the instability eradication method, the first step consists of detecting a set of conditions including (i) a resonance condition assessed from the spectral parameters, (ii) a duration condition detected when the resonance condition has prevailed for at least the M most recent frames, M being an integer greater than 1, and (iii) a gain condition which evidences consistently-high values of the pitch gain in the N most recent frames, N being an integer greater than 1. To eradicate the occasional instability, the pitch gain is reduced to a value lower than a given threshold whenever these three conditions are detected.

Instability eradication for analysis-by-synthesis speech codecs

A method and device eradicate the occasional instability inherent in analysis-by-synthesis speech/audio codecs and caused in particular by channel errors during transmission of highly periodic signals such as high-frequency sine waves. Analysis-by-synthesis techniques involve production, in response to the speech/audio signal and at regular time intervals called frames, of (a) a set of spectral parameters for use in driving a synthesis filter in view of synthesizing the speech/audio signal, and (b) a pitch gain for constructing a past-excitation-signal component supplied to the synthesis filter. In accordance with the instability eradication method, the first step consists of detecting a set of conditions including (i) a resonance condition assessed from the spectral parameters, (ii) a duration condition detected when the resonance condition has prevailed for at least the M most recent frames, M being an integer greater than 1, and (iii) a gain condition which evidences consistently-high values of the pitch gain in the N most recent frames, N being an integer greater than 1. To eradicate the occasional instability, the pitch gain is reduced to a value lower than a given threshold whenever these three conditions are detected.

Method and device for eradicating instability due to periodic signals in

A codebook is searched in view of encoding a sound signal. This codebook consists of a set of pulse amplitude/position combinations each defining L different positions and comprising both zero-amplitude pulses and non-zero-amplitude pulses assigned to respective positions p=1, 2, . . . L of the combination, wherein each non-zero-amplitude pulse assumes at least one of q possible amplitudes. To reduce the search complexity, a subset of pulse amplitude/position combinations from the codebook is pre-selected in relation to the sound signal, and only this subset of combinations is searched. The pre-selection of the subset of combinations consists of pre-establishing, in relation to the sound signal, a function S.sub.p between the respective positions p=1, 2, . . . L and the q possible amplitudes, the search being limited to the combinations of the codebook having non-zero-amplitude pulses which respect the pre-established function. The function can be pre-established by pre-assigning one of the q possible amplitudes to each position p, the pre-established function being respected when the non-zero-amplitude pulses of a combination each have an amplitude equal to the amplitude S.sub.p pre-assigned to the position p of that pulse.

Algebraic codebook with signal-selected pulse amplitude/position

A codebook is searched in view of encoding a sound signal. This codebook consists of a set of codevectors each of 40 positions and comprising N non-zero-amplitude pulses assignable to predetermined valid positions. To reduce the search complexity, a depth-first search is used which involves a tree structure with levels ordered from 1 through M. A path-building operation takes place at each level whereby a candidate path from the previous level is extended by choosing a predetermined number of new pulses and selecting valid positions for said new pulses in accordance with a given pulse-order rule and a given selection criterion. A path originated at the first level and extended by the path-building operations of subsequent levels determines the respective positions of the N non-zero-amplitude pulse of a candidate codevector. Use of a signal-based pulse-position likelihood estimate during the first few levels enable initial pulse-screening to start the search on favorable conditions. A selection criterion based on maximizing a ratio is used to assess the progress and to choose the best one among competing candidate codevectors.

Depth-first algebraic-codebook search for fast coding of speech

A method of encoding a speech signal is provided. This method improves the excitation codebook and search procedure of the conventional Code-Excited Linear Prediction (CELP) speech encoders. This code is based on a sparse algebraic code consisting in particular, but not exclusively, of interleaving N single-pulse permutation codes. The search complexity in finding the best codeword is greatly reduced by bringing the search back to the algebraic code domain thereby allowing the sparsity of the algebraic code to speed up the necessary computations. More precisely, the sparsity of the code enable the use of a very fast procedure based on N-embedded computation loops.

Fast sparse-algebraic-codebook search for efficient speech coding

The present invention concerns efficient quantization of more than one LPC spectral models per frame in order to enhance the accuracy of the time-varying spectrum representation without compromising on the coding-rate. Such efficient representation of LPC spectral models is advantageous to a number of techniques used for digital encoding of speech and/or audio signals.

Growing community of inventors

Sherbrooke, Canada

Claude Laflamme