tPLCnet: Real-time Deep Packet Loss Concealment in the Time Domain Using a Short Temporal Context

TL;DR

This paper presents tPLCnet, a real-time deep neural network for packet loss concealment in speech communication, which predicts lost frames efficiently using a short temporal context and outperforms baseline methods in quality.

Contribution

The paper introduces a novel seq2one neural network architecture for real-time packet loss concealment that operates efficiently with minimal context and demonstrates competitive performance.

Findings

Achieves robust PLC performance with low complexity

Outperforms zero-filling baseline across all metrics

Secures 3rd place in the Audio PLC Challenge

Abstract

This paper introduces a real-time time-domain packet loss concealment (PLC) neural-network (tPLCnet). It efficiently predicts lost frames from a short context buffer in a sequence-to-one (seq2one) fashion. Because of its seq2one structure, a continuous inference of the model is not required since it can be triggered when packet loss is actually detected. It is trained on 64h of open-source speech data and packet-loss traces of real calls provided by the Audio PLC Challenge. The model with the lowest complexity described in this paper reaches a robust PLC performance and consistent improvements over the zero-filling baseline for all metrics. A configuration with higher complexity is submitted to the PLC Challenge and shows a performance increase of 1.07 compared to the zero-filling baseline in terms of PLC-MOS on the blind test set and reaches a competitive 3rd place in the challenge ranking.