Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization

H. Ohmuro, Takehiro Moriya, Kazunori Mano, Satoshi Miki

Research output: Contribution to conference › Paper

19 Citations (Scopus)

Abstract

This paper proposes a new efficient line spectrum pair (LSP) vector quantization method which uses moving average (MA) interframe prediction of the parameters. This method has two advantages over auto-regressive prediction: the degradation of decoded parameters caused by bit errors affects only a few of the following frames, and codeword decoding can start at any frame. Using MA prediction, the final reconstructed LSP vector (quantized vector) of the current frame is represented as a linear combination of current and previous frame code vectors. In this paper we describe ways to achieve a more efficient LSP coder when each frame has four 10-ms subfraines. The spectral distance obtained by this method at 30 bits per 40 ms is better than that obtained using conventional multi-stage VQ without interframe correlation at 40 bits per 40 ms.

Original languageEnglish
Pages63-64
Number of pages2
DOIs
Publication statusPublished - 1993 Jan 1
Externally publishedYes
Event1993 IEEE Workshop on Speech Coding for Telecommunications: Speech Coding for the Network of the Future, SCFT 1993 - Sainte-Adele, Canada
Duration: 1993 Oct 13 → 1993 Oct 15

Conference

Conference1993 IEEE Workshop on Speech Coding for Telecommunications: Speech Coding for the Network of the Future, SCFT 1993
CountryCanada
CitySainte-Adele
Period93/10/13 → 93/10/15

Fingerprint

Vector quantization
coding
Decoding
Degradation

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Communication
  • Computer Networks and Communications

Cite this

Ohmuro, H., Moriya, T., Mano, K., & Miki, S. (1993). Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization. 63-64. Paper presented at 1993 IEEE Workshop on Speech Coding for Telecommunications: Speech Coding for the Network of the Future, SCFT 1993, Sainte-Adele, Canada. https://doi.org/10.1109/SCFT.1993.762340

Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization. / Ohmuro, H.; Moriya, Takehiro; Mano, Kazunori; Miki, Satoshi.

1993. 63-64 Paper presented at 1993 IEEE Workshop on Speech Coding for Telecommunications: Speech Coding for the Network of the Future, SCFT 1993, Sainte-Adele, Canada.

Research output: Contribution to conference › Paper

Ohmuro, H, Moriya, T, Mano, K & Miki, S 1993, 'Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization', Paper presented at 1993 IEEE Workshop on Speech Coding for Telecommunications: Speech Coding for the Network of the Future, SCFT 1993, Sainte-Adele, Canada, 93/10/13 - 93/10/15 pp. 63-64. https://doi.org/10.1109/SCFT.1993.762340
Ohmuro H, Moriya T, Mano K, Miki S. Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization. 1993. Paper presented at 1993 IEEE Workshop on Speech Coding for Telecommunications: Speech Coding for the Network of the Future, SCFT 1993, Sainte-Adele, Canada. https://doi.org/10.1109/SCFT.1993.762340
Ohmuro, H. ; Moriya, Takehiro ; Mano, Kazunori ; Miki, Satoshi. / Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization. Paper presented at 1993 IEEE Workshop on Speech Coding for Telecommunications: Speech Coding for the Network of the Future, SCFT 1993, Sainte-Adele, Canada.2 p.
@conference{8577548d16424c398d28c40bc9a5fc4b,
title = "Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization",
abstract = "This paper proposes a new efficient line spectrum pair (LSP) vector quantization method which uses moving average (MA) interframe prediction of the parameters. This method has two advantages over auto-regressive prediction: the degradation of decoded parameters caused by bit errors affects only a few of the following frames, and codeword decoding can start at any frame. Using MA prediction, the final reconstructed LSP vector (quantized vector) of the current frame is represented as a linear combination of current and previous frame code vectors. In this paper we describe ways to achieve a more efficient LSP coder when each frame has four 10-ms subfraines. The spectral distance obtained by this method at 30 bits per 40 ms is better than that obtained using conventional multi-stage VQ without interframe correlation at 40 bits per 40 ms.",
author = "H. Ohmuro and Takehiro Moriya and Kazunori Mano and Satoshi Miki",
year = "1993",
month = "1",
day = "1",
doi = "10.1109/SCFT.1993.762340",
language = "English",
pages = "63--64",
note = "1993 IEEE Workshop on Speech Coding for Telecommunications: Speech Coding for the Network of the Future, SCFT 1993 ; Conference date: 13-10-1993 Through 15-10-1993",

}

TY - CONF

T1 - Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization

AU - Ohmuro, H.

AU - Moriya, Takehiro

AU - Mano, Kazunori

AU - Miki, Satoshi

PY - 1993/1/1

Y1 - 1993/1/1

N2 - This paper proposes a new efficient line spectrum pair (LSP) vector quantization method which uses moving average (MA) interframe prediction of the parameters. This method has two advantages over auto-regressive prediction: the degradation of decoded parameters caused by bit errors affects only a few of the following frames, and codeword decoding can start at any frame. Using MA prediction, the final reconstructed LSP vector (quantized vector) of the current frame is represented as a linear combination of current and previous frame code vectors. In this paper we describe ways to achieve a more efficient LSP coder when each frame has four 10-ms subfraines. The spectral distance obtained by this method at 30 bits per 40 ms is better than that obtained using conventional multi-stage VQ without interframe correlation at 40 bits per 40 ms.

AB - This paper proposes a new efficient line spectrum pair (LSP) vector quantization method which uses moving average (MA) interframe prediction of the parameters. This method has two advantages over auto-regressive prediction: the degradation of decoded parameters caused by bit errors affects only a few of the following frames, and codeword decoding can start at any frame. Using MA prediction, the final reconstructed LSP vector (quantized vector) of the current frame is represented as a linear combination of current and previous frame code vectors. In this paper we describe ways to achieve a more efficient LSP coder when each frame has four 10-ms subfraines. The spectral distance obtained by this method at 30 bits per 40 ms is better than that obtained using conventional multi-stage VQ without interframe correlation at 40 bits per 40 ms.

UR - http://www.scopus.com/inward/record.url?scp=84985847724&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84985847724&partnerID=8YFLogxK

U2 - 10.1109/SCFT.1993.762340

DO - 10.1109/SCFT.1993.762340

M3 - Paper

AN - SCOPUS:84985847724

SP - 63

EP - 64

ER -

Access to Document