English

Full-Reference Speech Quality Estimation with Attentional Siamese Neural Networks

Audio and Speech Processing 2021-05-04 v1 Artificial Intelligence Machine Learning Sound

Abstract

In this paper, we present a full-reference speech quality prediction model with a deep learning approach. The model determines a feature representation of the reference and the degraded signal through a siamese recurrent convolutional network that shares the weights for both signals as input. The resulting features are then used to align the signals with an attention mechanism and are finally combined to estimate the overall speech quality. The proposed network architecture represents a simple solution for the time-alignment problem that occurs for speech signals transmitted through Voice-Over-IP networks and shows how the clean reference signal can be incorporated into speech quality models that are based on end-to-end trained neural networks.

Keywords

Cite

@article{arxiv.2105.00783,
  title  = {Full-Reference Speech Quality Estimation with Attentional Siamese Neural Networks},
  author = {Gabriel Mittags and Sebastian Möller},
  journal= {arXiv preprint arXiv:2105.00783},
  year   = {2021}
}

Comments

Late upload, presented at ICASSP 2020

R2 v1 2026-06-24T01:43:40.630Z