In this work we present a formal procedure for automating the tuning of the various parameters comprising a voice quality enhancer. First, we formalize the problem of tuning as a large-scale nonlinear programming problem. Second, we evaluate the performance of perceptual objective quality measures as optimization criteria for our tuning problem. We then perform a subjective quality assessment to compare the output of a voice enhancer obtained with parameters calculated with these different criteria and also with those obtained through a conventional approach of tuning by expert listening. The results show that using this automated methodology performs well in finding reasonable solutions for the tuning problem, potentially saving time and resources over manual evaluation and tuning.
Authors:
Giacobello, Daniele; Atkins, Joshua; Wung, Jason; Prabhu, Raghavendra
Affiliations:
Beats By Dr. Dre, Santa Monica, CA, USA; Beats Electronics, LLC, Santa Monica, CA, USA; Beats by Dr. Dre, Santa Monica, CA, USA(See document for exact affiliation information.)
AES Convention:
135 (October 2013)
eBrief:114
Publication Date:
October 16, 2013
Click to purchase paper as a non-member or you can login as an AES member to see more options.
No AES members have commented on this paper yet.
To be notified of new comments on this paper you can subscribe to this RSS feed. Forum users should login to see additional options.
If you are not yet an AES member and have something important to say about this paper then we urge you to join the AES today and make your voice heard. You can join online today by clicking here.