DocumentCode
730805
Title
A novel sinusoidal approach to audio signal frame loss concealment and its application in the new evs codec standard
Author
Bruhn, Stefan ; Norvell, Erik ; Svedberg, Jonas ; Sverrisson, Sigurdur
Author_Institution
SMN, Ericsson AB, Stockholm, Sweden
fYear
2015
fDate
19-24 April 2015
Firstpage
5142
Lastpage
5146
Abstract
The new 3GPP codec for Enhanced Voice Services (EVS) comprises a collection of frame loss concealment techniques, each specifically designed for the different coding modes of that codec. One of them, called “Phase Error Concealment Unit (Phase ECU)”, was developed for the High Quality (HQ) MDCT coding mode. Despite this target application, Phase ECU is a generic stand-alone tool operating on a buffer of the previously decoded and reconstructed time signal. Its framework is based on the sinusoidal analysis and synthesis paradigm. Besides a description of the basic technology we present optimizations and adaptations required for meeting the challenging 3GPP EVS codec performance requirements, and that make the method robust for a broad range of audio signals under various frame loss conditions from isolated frame erasures to severe burst loss. Test results are reported that show significant improvements over traditional techniques.
Keywords
audio coding; codecs; 3GPP codec; EVS codec standard; Enhanced Voice Services; High Quality MDCT coding mode; Phase ECU; Phase Error Concealment Unit; audio signal frame loss concealment; audio signals; coding modes; decoded time signal; frame loss concealment techniques; frame loss conditions; isolated frame erasures; reconstructed time signal; severe burst loss; sinusoidal analysis; sinusoidal synthesis paradigm; Discrete Fourier transforms; Encoding; Prototypes; Speech; Speech codecs; Transient analysis; EVS; error concealment; speech/audio coding;
fLanguage
English
Publisher
ieee
Conference_Titel
Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on
Conference_Location
South Brisbane, QLD
Type
conf
DOI
10.1109/ICASSP.2015.7178951
Filename
7178951
Link To Document