Generalized predictive trellis-coded quantization of speech

Michael W Marcellin, Thomas R. Fischer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Trellis-coded quantization (TCQ) is incorporated into a generalized predicted (noise feedback) structure for encoding sampled speech. Adaptive residual encoding and adaptive prediction are used to obtain signal-to-noise ratios (SNR) in the range of 17.5 to 21.5 dB for encoding sampled speech with 2 bits/sample (16 kb/s). Spectral noise shaping is achieved with this coding structure by using a bandwidth-expanded version of the predictor as a noise feedback filter. This has the effect of reducing the SNR of the encoding by a small amount but improving the perceptual quality of the reconstructed speech. The result is of excellent communications quality. A modified Viterbi algorithm is used for the trellis search to allow flexibility in the choice of symbol release rule. The effects of varying the encoding delay and the number of symbols released per trace-back on system performance and complexity are investigated. It is shown that excellent SNR performance can be obtained with modest encoding delays.

Original languageEnglish (US)
Title of host publicationEighth Annu Int Phoenix Conf Comput Commun 1989 Conf Proc
Editors Anon
PublisherPubl by IEEE
Pages144-148
Number of pages5
StatePublished - 1989
EventEighth Annual International Phoenix Conference on Computers and Communications - 1989 Conference Proceedings - Scottsdale, AZ, USA
Duration: Mar 22 1989Mar 24 1989

Other

OtherEighth Annual International Phoenix Conference on Computers and Communications - 1989 Conference Proceedings
CityScottsdale, AZ, USA
Period3/22/893/24/89

Fingerprint

Signal to noise ratio
Feedback
Viterbi algorithm
Bandwidth
Communication

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Marcellin, M. W., & Fischer, T. R. (1989). Generalized predictive trellis-coded quantization of speech. In Anon (Ed.), Eighth Annu Int Phoenix Conf Comput Commun 1989 Conf Proc (pp. 144-148). Publ by IEEE.

Generalized predictive trellis-coded quantization of speech. / Marcellin, Michael W; Fischer, Thomas R.

Eighth Annu Int Phoenix Conf Comput Commun 1989 Conf Proc. ed. / Anon. Publ by IEEE, 1989. p. 144-148.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Marcellin, MW & Fischer, TR 1989, Generalized predictive trellis-coded quantization of speech. in Anon (ed.), Eighth Annu Int Phoenix Conf Comput Commun 1989 Conf Proc. Publ by IEEE, pp. 144-148, Eighth Annual International Phoenix Conference on Computers and Communications - 1989 Conference Proceedings, Scottsdale, AZ, USA, 3/22/89.
Marcellin MW, Fischer TR. Generalized predictive trellis-coded quantization of speech. In Anon, editor, Eighth Annu Int Phoenix Conf Comput Commun 1989 Conf Proc. Publ by IEEE. 1989. p. 144-148
Marcellin, Michael W ; Fischer, Thomas R. / Generalized predictive trellis-coded quantization of speech. Eighth Annu Int Phoenix Conf Comput Commun 1989 Conf Proc. editor / Anon. Publ by IEEE, 1989. pp. 144-148
@inproceedings{af11485d24fb4a8da7f09ecee73ce749,
title = "Generalized predictive trellis-coded quantization of speech",
abstract = "Trellis-coded quantization (TCQ) is incorporated into a generalized predicted (noise feedback) structure for encoding sampled speech. Adaptive residual encoding and adaptive prediction are used to obtain signal-to-noise ratios (SNR) in the range of 17.5 to 21.5 dB for encoding sampled speech with 2 bits/sample (16 kb/s). Spectral noise shaping is achieved with this coding structure by using a bandwidth-expanded version of the predictor as a noise feedback filter. This has the effect of reducing the SNR of the encoding by a small amount but improving the perceptual quality of the reconstructed speech. The result is of excellent communications quality. A modified Viterbi algorithm is used for the trellis search to allow flexibility in the choice of symbol release rule. The effects of varying the encoding delay and the number of symbols released per trace-back on system performance and complexity are investigated. It is shown that excellent SNR performance can be obtained with modest encoding delays.",
author = "Marcellin, {Michael W} and Fischer, {Thomas R.}",
year = "1989",
language = "English (US)",
pages = "144--148",
editor = "Anon",
booktitle = "Eighth Annu Int Phoenix Conf Comput Commun 1989 Conf Proc",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Generalized predictive trellis-coded quantization of speech

AU - Marcellin, Michael W

AU - Fischer, Thomas R.

PY - 1989

Y1 - 1989

N2 - Trellis-coded quantization (TCQ) is incorporated into a generalized predicted (noise feedback) structure for encoding sampled speech. Adaptive residual encoding and adaptive prediction are used to obtain signal-to-noise ratios (SNR) in the range of 17.5 to 21.5 dB for encoding sampled speech with 2 bits/sample (16 kb/s). Spectral noise shaping is achieved with this coding structure by using a bandwidth-expanded version of the predictor as a noise feedback filter. This has the effect of reducing the SNR of the encoding by a small amount but improving the perceptual quality of the reconstructed speech. The result is of excellent communications quality. A modified Viterbi algorithm is used for the trellis search to allow flexibility in the choice of symbol release rule. The effects of varying the encoding delay and the number of symbols released per trace-back on system performance and complexity are investigated. It is shown that excellent SNR performance can be obtained with modest encoding delays.

AB - Trellis-coded quantization (TCQ) is incorporated into a generalized predicted (noise feedback) structure for encoding sampled speech. Adaptive residual encoding and adaptive prediction are used to obtain signal-to-noise ratios (SNR) in the range of 17.5 to 21.5 dB for encoding sampled speech with 2 bits/sample (16 kb/s). Spectral noise shaping is achieved with this coding structure by using a bandwidth-expanded version of the predictor as a noise feedback filter. This has the effect of reducing the SNR of the encoding by a small amount but improving the perceptual quality of the reconstructed speech. The result is of excellent communications quality. A modified Viterbi algorithm is used for the trellis search to allow flexibility in the choice of symbol release rule. The effects of varying the encoding delay and the number of symbols released per trace-back on system performance and complexity are investigated. It is shown that excellent SNR performance can be obtained with modest encoding delays.

UR - http://www.scopus.com/inward/record.url?scp=0024905998&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024905998&partnerID=8YFLogxK

M3 - Conference contribution

SP - 144

EP - 148

BT - Eighth Annu Int Phoenix Conf Comput Commun 1989 Conf Proc

A2 - Anon, null

PB - Publ by IEEE

ER -