Recently software systems have been proposed to accelerate the progress of singing beginners. The basics of these systems are: the pitch of the sung notes is detected and algorithmic errors removed. Then, an alignment is made with a melodic ground truth, often as a midi representation, using techniques including Dynamic Time Warping and Hidden Markov Models. Although results have been reasonable, significant drawbacks to these alignment schemes include how a “musically acceptable” alignment can be identified, dynamic singer behavior, multiple repeated notes, and dealing with omitted or extra notes. To this end an improved singing analysis system structure is proposed that includes psychoacoustic models and intelligent decision making. Justification is given along with a description of a structured evaluation procedure.
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.
