We can remove the noise of other people talking in the background to help cochlear implant users understand speech in noise!

​Roozbeh Soleymani of the EAR Lab has developed a noise reduction algorithm named SEDA which is specifically designed to reduce "multi-talker babble" noise (i.e. the noise of other people talking in the background.  Average improvement for CI users was between 7 and 18 percentage points more words (depending on the signal to noise ratio) on the very difficult IEEE sentence test.  Additionally, all  CI users rated the sound quality of speech processed with our SEDA noise reduction as having a better sound quality than without it.

Speech intelligibility of IEEE sentences in multi-talker babble with (blue) and without (red) SEDA processing averaged across 7 subjects. The purple dashed line indicates performance for the subject in quiet, representing the best possible score for that subject.

​For more information, check out the article (downloadable below) or the SEDA webpage.

Soleymani, R., Selesnick, I. W., and Landsberger, D. M. (2018). "SEDA: A tunable Q-factor wavelet-based noise reduction algorithm for multi-talker babble," Speech Communication 96, 102-115. pdf