BS-PLCNet 2: Two-stage Band-split Packet Loss Concealment Network with Intra-model Knowledge Distillation

Zihan Zhang, Xianjun Xia, Chuanzeng Huang, Yijian Xiao, Lei Xie

Research output: Contribution to journalConference articlepeer-review

Abstract

Audio packet loss is an inevitable problem in real-time speech communication.A band-split packet loss concealment network (BS-PLCNet) targeting full-band signals was recently proposed.Although it performs superiorly in the ICASSP 2024 PLC Challenge, BS-PLCNet is a large model with high computational complexity of 8.95G FLOPS.This paper presents its updated version, BS-PLCNet 2, to reduce computational complexity and improve performance further.Specifically, to compensate for the missing future information, in the wide-band module, we design a dual-path encoder structure (with non-causal and causal path) and leverage an intra-model knowledge distillation strategy to distill the future information from the non-causal teacher to the casual student.Moreover, we introduce a lightweight post-processing module after packet loss restoration to recover speech distortions and remove residual noise in the audio signal.With only 40% of original parameters in BS-PLCNet, BS-PLCNet 2 brings 0.18 PLCMOS improvement on the ICASSP 2024 PLC challenge blind set, achieving state-of-the-art performance on this dataset.

Original languageEnglish
Pages (from-to)1750-1754
Number of pages5
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
DOIs
StatePublished - 2024
Event25th Interspeech Conferece 2024 - Kos Island, Greece
Duration: 1 Sep 20245 Sep 2024

Keywords

  • band-split
  • intra-model knowledge distillation
  • packet loss concealment
  • two-stage

Fingerprint

Dive into the research topics of 'BS-PLCNet 2: Two-stage Band-split Packet Loss Concealment Network with Intra-model Knowledge Distillation'. Together they form a unique fingerprint.

Cite this