Fast Randomization for Distributed Low-Bitrate Coding of Speech and Audio

Tutkimustuotos: Lehtiartikkeli

Standard

Fast Randomization for Distributed Low-Bitrate Coding of Speech and Audio. / Backstrom, Tom; Fischer, Johannes.

julkaisussa: IEEE/ACM Transactions on Audio Speech and Language Processing, Vuosikerta 26, Nro 1, 01.2018, s. 19-30.

Tutkimustuotos: Lehtiartikkeli

Harvard

APA

Vancouver

Author

Bibtex - Lataa

@article{238136e58df94d92a9e84edebba32aa6,
title = "Fast Randomization for Distributed Low-Bitrate Coding of Speech and Audio",
abstract = "Efficient coding of speech and audio in a distributed system requires that quantization errors across nodes are uncorrelated. Yet with conventional methods at low bitrates, quantization levels become increasingly sparse, which does not correspond to the distribution of the input signal and importantly, also reduces coding efficiency in a distributed system. We have recently proposed a distributed speech and audio codec design which applies quantization in a randomized domain such that quantization errors are randomly rotated in the output domain. Similar to dithering, this ensures that quantization errors across nodes are uncorrelated and coding efficiency is retained. In this paper we improve this approach by proposing faster randomization methods, with a computational complexity O(N log N). Presented experiments demonstrate that the proposed randomizations yield uncorrelated signals, that perceptual quality is competitive and that the complexity of the proposed methods is feasible for practical applications.",
keywords = "audio coding, Codecs, Complexity theory, distributed coding, orthonormal matrix, Quantization (signal), randomization, Speech, Speech coding, speech coding, Speech processing, superfast algorithm",
author = "Tom Backstrom and Johannes Fischer",
year = "2018",
month = "1",
doi = "10.1109/TASLP.2017.2757601",
language = "English",
volume = "26",
pages = "19--30",
journal = "IEEE/ACM Transactions on Audio, Speech, and Language Processing",
issn = "2329-9290",
publisher = "IEEE Advancing Technology for Humanity",
number = "1",

}

RIS - Lataa

TY - JOUR

T1 - Fast Randomization for Distributed Low-Bitrate Coding of Speech and Audio

AU - Backstrom, Tom

AU - Fischer, Johannes

PY - 2018/1

Y1 - 2018/1

N2 - Efficient coding of speech and audio in a distributed system requires that quantization errors across nodes are uncorrelated. Yet with conventional methods at low bitrates, quantization levels become increasingly sparse, which does not correspond to the distribution of the input signal and importantly, also reduces coding efficiency in a distributed system. We have recently proposed a distributed speech and audio codec design which applies quantization in a randomized domain such that quantization errors are randomly rotated in the output domain. Similar to dithering, this ensures that quantization errors across nodes are uncorrelated and coding efficiency is retained. In this paper we improve this approach by proposing faster randomization methods, with a computational complexity O(N log N). Presented experiments demonstrate that the proposed randomizations yield uncorrelated signals, that perceptual quality is competitive and that the complexity of the proposed methods is feasible for practical applications.

AB - Efficient coding of speech and audio in a distributed system requires that quantization errors across nodes are uncorrelated. Yet with conventional methods at low bitrates, quantization levels become increasingly sparse, which does not correspond to the distribution of the input signal and importantly, also reduces coding efficiency in a distributed system. We have recently proposed a distributed speech and audio codec design which applies quantization in a randomized domain such that quantization errors are randomly rotated in the output domain. Similar to dithering, this ensures that quantization errors across nodes are uncorrelated and coding efficiency is retained. In this paper we improve this approach by proposing faster randomization methods, with a computational complexity O(N log N). Presented experiments demonstrate that the proposed randomizations yield uncorrelated signals, that perceptual quality is competitive and that the complexity of the proposed methods is feasible for practical applications.

KW - audio coding

KW - Codecs

KW - Complexity theory

KW - distributed coding

KW - orthonormal matrix

KW - Quantization (signal)

KW - randomization

KW - Speech

KW - Speech coding

KW - speech coding

KW - Speech processing

KW - superfast algorithm

UR - http://www.scopus.com/inward/record.url?scp=85030764240&partnerID=8YFLogxK

U2 - 10.1109/TASLP.2017.2757601

DO - 10.1109/TASLP.2017.2757601

M3 - Article

VL - 26

SP - 19

EP - 30

JO - IEEE/ACM Transactions on Audio, Speech, and Language Processing

JF - IEEE/ACM Transactions on Audio, Speech, and Language Processing

SN - 2329-9290

IS - 1

ER -

ID: 15870224