Crowdsourcing approach for subjective evaluation of echo impairment

TL;DR

This paper introduces a crowdsourcing method for subjective evaluation of echo impairment in real-time communication, demonstrating its accuracy, reproducibility, and utility in a challenge setting, addressing limitations of traditional objective metrics.

Contribution

The paper presents an open-source crowdsourcing approach for subjective echo impairment evaluation, providing a scalable, cost-effective alternative to lab-based tests and objective metrics.

Findings

Crowdsourcing evaluation correlates well with subjective assessments.

The tool is highly reproducible across different users.

It enabled rapid, cost-effective evaluation in the ICASSP 2021 AEC Challenge.

Abstract

The quality of acoustic echo cancellers (AECs) in real-time communication systems is typically evaluated using objective metrics like ERLE and PESQ, and less commonly with lab-based subjective tests like ITU-T Rec. P.831. We will show that these objective measures are not well correlated to subjective measures. We then introduce an open-source crowdsourcing approach for subjective evaluation of echo impairment which can be used to evaluate the performance of AECs. We provide a study that shows this tool is accurate and highly reproducible. This new tool has been recently used in the ICASSP 2021 AEC Challenge which made the challenge possible to do quickly and cost effectively.