SDR: 18.635PESQ: 2.436ΔPESQ: 1.16
Papers with Code · CC BY-SA 4.0
Speech Enhancement
Speech Enhancement is a signal processing task that involves improving the quality of speech signals captured under noisy or degraded conditions. The goal of speech enhancement is to make speech signals clearer, more intelligible, and more pleasant to listen to, which can be used for various applications such as voice recognition, teleconferencing, and hearing aids. A representative Github project with online demo : ClearerVoice-Studio.
<span style="color:grey; opacity: 0.6">( Image credit: A Fully Convolutional Neural Network For Speech Enhancement )</span>
논문
Audio Quality MOS: 4.19
PESQ-NB: 3.21
Face Landmark-based Speaker-Independent Audio-Visual Speech Enhancement in Multi-Talker Environments
PESQ: 3.03
PESQ-WB: 2.75
PESQ-WB: 2.10DNSMOS OVRL: 3.25DNSMOS: 3.82DNSMOS BAK: 4.03
PESQ-WB: 1.60SI-SDR: 8.35ESTOI: 0.53SIGMOS: 2.08
PESQ-WB: 2.31SI-SDR: 16.93ESTOI: 0.70SIGMOS: 2.69
SDR: -4.76
STOI: 0.544PESQ: 1.17ViSQOL: 1.68HASQI: 0.249
PESQ: 3.07SDR: 15.04SI-SNR: 14.35
PESQ-WB: 2.37SI-SDR: 16.92ESTOI: 0.71SIGMOS: 2.87
CBAK: 3.24CSIG: 3.52PESQ: 2.4COVL: 2.96
SDR: 19.67PESQ: 2.67STOI: 0.973
PESQ-WB: 3.09SI-SDR: 16.29ESTOI: 0.73SIGMOS: 3.18
PESQ: 3.99SI-SDR: 0.40PESQ (wb): 3.99ESTOI: 0.83
PESQ-WB: 2.33SI-SDR: 17.85ESTOI: 0.73SIGMOS: 3.44
PESQ-WB: 2.50SI-SDR: 16.78ESTOI: 0.73SIGMOS: 3.41