Binary Pulse Excitation: A Novel Approach to Low Complexity CELP Coding

The last decade has witnessed an acceleration in the evolution of speech coding. Near-toll quality speech coders are now available at bit rates from 4.8 to 8 kbit/s. These low bit rate speech coders are becoming increasingly needed for many future applications such as digital mobile radio telephony, mobile satellite links, and the emerging ISDN service. High quality speech at these low bit rates has become possible with the introduction of a new generation of speech coding techniques known as analysis-by-synthesis predictive coding. The structure of this new generation of speech coders first appeared with the introduction of multi-pulse excited linear prediction coder (MPE-LPC) by Atal in 1982 [1]. Since its introduction, the MPE has received much attention from researchers, and as a result, several analysis-by-synthesis coders have been developed in the bit rate range from 4.8 to 16 kbit/s with different levels of complexity [2]. These systems include the regular-pulse excited LPC (RPE-LPC) [3], the code-excited LPC (CELP) [4], and the self-excited LPC [5] (or Backward Excitation Recovery (BER) [6]). All the above mentioned coders exhibit the same structure as the originally proposed MPE-LPC [1] in which the excitation signal is optimized by minimizing the perceptually weighted error between the original and synthesized speech. They differ only in the way the excitation signal is defined and coded.

[1]  Manfred R. Schroeder,et al.  Code-excited linear prediction(CELP): High-quality speech at very low bit rates , 1985, ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  M. Delprat,et al.  RPCELP: a high quality and low complexity scheme for narrow band coding of speech , 1988, 8th European Conference on Electrotechnics, Conference Proceedings on Area Communication.

[3]  R. Steele,et al.  Performance of error protected binary pulse excitation coders at 11.4 kb/s over mobile radio channels , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[4]  Ed F. Deprettere,et al.  A class of analysis-by-synthesis predictive coders for high quality speech coding at rates between 4.8 and 16 kbit/s , 1988, IEEE J. Sel. Areas Commun..

[5]  B. Atal,et al.  Predictive coding of speech signals and subjective error criteria , 1979 .

[6]  S. Morissette,et al.  On reducing computational complexity of codebook search in CELP coder through the use of algebraic codes , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[7]  P. Mabilleau,et al.  Fast CELP coding based on algebraic codes , 1987, ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[8]  Thomas P. Barnwell,et al.  Quality comparison of low complexity 4800 bps self excited and code excited vocoders , 1987, ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[9]  Alain Le Guyader,et al.  A robust and fast celp coder at 16 kbit/s , 1988, Speech Commun..

[10]  Ed F. Deprettere,et al.  Regular-pulse excitation-A novel approach to effective and efficient multipulse coding of speech , 1986, IEEE Trans. Acoust. Speech Signal Process..

[11]  W. Bastiaan Kleijn,et al.  An efficient stochastically excited linear predictive coding algorithm for high quality low bit rate transmission of speech , 1988, Speech Commun..

[12]  Bishnu S. Atal,et al.  A new model of LPC excitation for producing natural-sounding speech at low bit rates , 1982, ICASSP.

[13]  C. Xydeas,et al.  Advances in analysis by synthesis LPC speech coders , 1987 .

[14]  Ira Alan Gerson,et al.  Vector Sum Excited Linear Prediction (VSELP) , 1991 .