Harmonic model for MDCT based audio coding with LPC envelope

Takehiro Moriya, Yutaka Kamamoto, Noboru Harada, Tom Backstrom, Christian Helmrich, Guillaume Fuchs

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

2 Citations (Scopus)

Abstract

Conventional music coders, based on a modified discrete cosine transform (MDCT) suffer greatly when lowering their bit-rate and delay. In particular, tonal music signals are penalized by short analysis windows and the variable length coding of the quantized MDCT coefficients demands a significant amount of bits for coding the harmonic structure. For solving such an issue, the paper proposes a frequency-domain harmonic model aiming to amend the probability model of the variable length coding of the quantized MDCT coefficients. The new model was combined successfully with an envelope based arithmetic coding at rate lower than 10 kbps, and with a context based arithmetic coding at higher bit rates in the recent 3 GPP EVS (Enhanced Voice Services) codec standard. Objective and subjective quality tests indicate that the proposed harmonic model enhances the quality of music for low-delay audio coding.

Original languageEnglish
Title of host publication2015 23rd European Signal Processing Conference, EUSIPCO 2015
PublisherIEEE
Pages789-793
Number of pages5
ISBN (Electronic)9780992862633
DOIs
Publication statusPublished - 22 Dec 2015
MoE publication typeA4 Article in a conference publication
EventEuropean Signal Processing Conference - Nice, France
Duration: 31 Aug 20154 Sep 2015
Conference number: 23

Conference

ConferenceEuropean Signal Processing Conference
Abbreviated titleEUSIPCO
CountryFrance
CityNice
Period31/08/201504/09/2015

Keywords

  • arithmetic coding
  • envelope
  • EVS
  • harmonic interval
  • MDCT

Fingerprint

Dive into the research topics of 'Harmonic model for MDCT based audio coding with LPC envelope'. Together they form a unique fingerprint.

Cite this