tPLCnet: Real-time Deep Packet Loss Concealment in the Time Domain Using a Short Temporal Context (2022-04-04T00:00:00.000000Z)

TL;DR

A real-time time-domain packet loss concealment (PLC) neural-network (tPLCnet) that efficiently predicts lost frames from a short context buffer in a sequence-to-one (seq2one) fashion.

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.

Authors

Nils L. Westhausen

2 papers

B. Meyer

2 papers

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

TL;DR

Abstract

Authors

References28 items

INTERSPEECH 2022 Audio Deep Packet Loss Concealment Challenge

Effect of Noise Suppression Losses on Speech Distortion and ASR Performance

A temporal-spectral generative adversarial network based end-to-end packet loss concealment for wideband speech transmission.

A Time-Domain Convolutional Recurrent Network for Packet Loss Concealment

Combining Adaptive Filtering And Complex-Valued Deep Postfiltering For Acoustic Echo Cancellation

Deep Noise Suppression with Non-Intrusive PESQNet Supervision Enabling the Use of Real Training Data

Low-Complexity, Real-Time Joint Neural Echo Control and Speech Enhancement Based On Percepnet

Towards Efficient Models for Real-Time Deep Noise Suppression

Dual-Path RNN for Long Recording Speech Separation

Dnsmos: A Non-Intrusive Perceptual Objective Speech Quality Metric to Evaluate Noise Suppressors

Acoustic Echo Cancellation with the Dual-Signal Transformation LSTM Network

A consolidated view of loss functions for supervised deep learning-based speech enhancement

ICASSP 2021 Acoustic Echo Cancellation Challenge: Datasets, Testing Framework, and Results

Data Augmentation and Loss Normalization for Deep Noise Suppression

A Perceptually-Motivated Approach for Low-Complexity, Real-Time Enhancement of Fullband Speech

The INTERSPEECH 2020 Deep Noise Suppression Challenge: Datasets, Subjective Testing Framework, and Challenge Results

Dual-Signal Transformation LSTM Network for Real-Time Noise Suppression

ConcealNet: An End-to-end Neural Network for Packet Loss Concealment in Deep Speech Emotion Recognition

Speech Loss Compensation by Generative Adversarial Networks

Vision-Infused Deep Audio Inpainting

Conv-TasNet: Surpassing Ideal Time–Frequency Magnitude Masking for Speech Separation

Exploring Tradeoffs in Models for Low-Latency Speech Enhancement

Speech Prediction Using an Adaptive Recurrent Neural Network with Application to Packet Loss Concealment

TaSNet: Time-Domain Audio Separation Network for Real-Time, Single-Channel Speech Separation

Tacotron: Towards End-to-End Speech Synthesis

Packet Loss Concealment Based on Deep Neural Networks for Digital Speech Transmission

Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling

Methods for objective and subjective assessment of quality Perceptual evaluation of speech quality ( PESQ ) : An objective method for end-to-end speech quality assessment of narrow-band telephone networks and speech codecs

Field of Study

Venue Information

Name

Type

URL

Alternate Names