Speech Coding Algorithms

Chu, Wai C.

doi:10.1002/0471668850

Cited by 260 publications

(94 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The order of the short-term and long-term predictors are respectively Nstp = 12 and NLT P = 1, obtained with the procedure of Section 3. or equivalently pitch frequency in the interval [82Hz, 571Hz]. The prediction residual vector is encoded according to [5] using 40 nonzero samples constrained with ±1 values and a gain. In the classical and optimized ACELP scheme, the order of the short-term and longterm analysis are the same (N f = 12 and Np = 1).…”

Section: Methodsmentioning

confidence: 99%

“…The coefficients of the short-term filter are found using the autocorrelation method on a subframe basis of 80 samples. The pitch delay and gain are found on the residual error signal according to traditional ACELP encoding [5]. The final residual error signal is also encoded according to [5] but on the subsamples frame basis with 20 non-zero samples and a gain that is averaged with the next one.…”

Section: Methodsmentioning

confidence: 99%

“…The final residual error signal is also encoded according to [5] but on the subsamples frame basis with 20 non-zero samples and a gain that is averaged with the next one. In order to obtain the same number of parameters for both regularized and traditional ACELP, the values obtained with regularized ACELP are being interpolated (the short-term filter interpolation is done in the LSF domain [5]), so that for each n-th subframe of each method, the transfer function is:…”

Section: Methodsmentioning

confidence: 99%

“…The final step is to encode the residual error signal after the two predictors that is hoped to be white and Gaussian. The encoding of the residual signal uses very few bits: in ACELP coders usually the residual is encoded with only 20-30% of non-zeros samples with constrained values of ±1 and a gain gac(n) [5].…”

Section: General Formulation For Linear Predictorsmentioning

confidence: 99%

See 3 more Smart Citations

Joint estimation of short-term and long-term predictors in speech coders

Giacobello

Christensen

Dahl

et al. 2009

2009 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

In low bit-rate coders, the near-sample and far-sample redundancies of the speech signal are usually removed by a cascade of a shortterm and a long-term linear predictor. These two predictors are usually found in a sequential and therefore suboptimal approach. In this paper we propose an analysis model that jointly finds the two predictors by adding a regularization term in the minimization process to impose sparsity constraints on a high order predictor. The result is a linear predictor that can be easily factorized into the short-term and long-term predictors. This estimation method is then incorporated into an Algebraic Code Excited Linear Prediction scheme and shows to have a better performance than traditional cascade methods and other joint optimization methods, offering lower distortion and higher perceptual speech quality.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: General Formulation For Linear Predictorsmentioning

confidence: 99%

See 2 more Smart Citations

Joint estimation of short-term and long-term predictors in speech coders

Giacobello

Christensen

Dahl

et al. 2009

2009 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

show abstract

“…In other words, an LSF of a certain frequency value affects mainly the PSD at the same frequency value. It is known as localisation property (Chu, 2003), where the modification to PSD have a local effect on the LSF. This is its advantage over other representation like LPCC, Reflection Coefficient (RC), LAR where changes in particular parameter affect the whole spectrum.…”

Section: Line Spectral Pairs Frequencies (Lsf)mentioning

confidence: 99%

On the use of perceptual Line Spectral pairs Frequencies and higher-order residual moments for Speaker Identification

Sahidullah¹,

Chakroborty²,

Saha³

2010

IJBM

View full text Add to dashboard Cite

Conventional Speaker Identification (SI) systems utilise spectral features like Mel-Frequency Cepstral Coefficients (MFCC) or PerceptualLinear Prediction (PLP) as a frontend module. Line Spectral pairs Frequencies (LSF) are popular alternative representation of Linear Prediction Coefficients (LPC). In this paper, an investigation is carried out to extract LSF from perceptually modified speech. A new feature set extracted from the residual signal is also proposed. SI system based on this residual feature containing complementary information to spectral characteristics, when fused with the conventional spectral feature based system as well as the proposed perceptually modified LSF, shows improved performance.Keywords: SI; speaker identification; LSF; line spectral pairs frequencies; perceptual linear prediction; residual signal; higher order statistics.Reference to this paper should be made as follows: Sahidullah, M., Chakroborty, S. and Saha, G. (2010)

show abstract