CITATION — REFERENCE ENTRY
An on-line algorithm for real-time accompaniment — Proceedings of the 1984 International Computer Music Conference
- Key
- dannenberg1984-online-accompaniment
- Authors
- Dannenberg, Roger B.
- Issued
- 1985
- Type
- paper-conference
- Container
- Proceedings of the 1984 International Computer Music Conference
- Publisher
- International Computer Music Association
- Pages
- 193-198
Raw CSL JSON
{
"URL": "https://www.cs.cmu.edu/~rbd/papers/icmc84accomp.pdf",
"page": "193-198",
"type": "paper-conference",
"title": "An on-line algorithm for real-time accompaniment",
"author": [
{
"given": "Roger B.",
"family": "Dannenberg"
}
],
"issued": {
"date-parts": [
[
1985
]
]
},
"publisher": "International Computer Music Association",
"container-title": "Proceedings of the 1984 International Computer Music Conference"
}
Claims
-
Real-time accompaniment avoids the synchronization problems of taped accompaniment but requires the computer to follow the soloist. Dannenberg divided this into three subproblems: detecting the soloist's input, matching it against a score of expected input while tolerating mistakes, and generating an accompaniment that follows the soloist. He presented an efficient dynamic programming algorithm, motivated by time-warping in speech recognition, that defines the best match as the longest common subsequence of the performance and the score.
"Real-time accompaniment solves most of the synchronization problems inherent in taped accompaniment; however, this new approach requires the computer to have the ability to follow the soloist. Three subproblems arise: detecting and processing input from the live performer, matching this input against a score of expected input, and generating the timing information necessary to control the generation of the accompaniment. It is expected that the live solo performance will contain mistakes or be imperfectly detected by the computer [...]. An efficient dynamic programming algorithm for finding the best match between solo performance and the score is presented. [...] The dynamic programming technique [10, 11] used for "time-warping" in speech research motivated the algorithm described in this paper. [...] The best match is the longest common subsequence of the two streams."
Available in