Improving performance of real-time full-band blind packet-loss concealment with predictive network
arXiv:2211.04071 · doi:10.1109/ICASSP49357.2023.10097132
Abstract
Packet loss concealment (PLC) is a tool for enhancing speech degradation caused by poor network conditions or underflow/overflow in audio processing pipelines. We propose a real-time recurrent method that leverages previous outputs to mitigate artefact of lost packets without the prior knowledge of loss mask. The proposed full-band recurrent network (FRN) model operates at 48 kHz, which is suitable for high-quality telecommunication applications. Experiment results highlight the superiority of FRN over an offline non-causal baseline and a top performer in a recent PLC challenge.
arXiv commentIn Proceedings ICASSP 2023, 5 pages, 1 figure, 4 tables
References in corpus (4)
- Sequence Transduction with Recurrent Neural Networks
- Looking to Listen at the Cocktail Party: A Speaker-Independent Audio-Visual Model for Speech Separation
- Deep speech inpainting of time-frequency masks
- TUNet: A Block-online Bandwidth Extension Model based on Transformers and Self-supervised Pretraining