We present the Bayesian Adaptive Speech Intelligibility Estimation (BASIE) method - a tool for rapid estimation of a given speech reception threshold (SRT) and the slope at that threshold of multiple psychometric functions for speech intelligibility in noise. The core of this tool is an adaptive Bayesian procedure, which adjusts the signal-to-noise ratio at each subsequent stimulus such that the expected variance of the threshold and slope estimates are minimised. Simulation results show that the algorithm is able to achieve SRT estimates accurate to within +-1 dB in under 30 iterations. Furthermore, we discuss strategies for using BASIE to evaluate the effects of speech processing algorithms on intelligibility and we give two illustrative examples for different noise reduction methods with supporting listening experiments.

Authors: Gaubitch, Nikolay D.; Brookes, Mike; Naylor, Patrick A.; Sharma, Dushyant
Affiliation: Imperial College London, UK
AES Conference: 39th International Conference: Audio Forensics: Practices and Challenges (June 2010)
Paper Number: 7-1
Publication Date: June 17, 2010 Import into BibTeX
Subject: Audio Forensics: Enhancement of Noisy Recordings
Permalink: https://www.aes.org/e-lib/browse.cfm?elib=15486

Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!

This paper costs $33 for non-members and is free for AES members and E-Library subscribers.

Learn more about the AES E-Library

E-Library Location: (CD 39thPapers)   /39/aes39-000010.pdf

Start a discussion about this paper!