Timbre Discrimination for Brief Instrument Sounds

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

24 Downloads (Pure)


Timbre discrimination, even for very brief sounds, allows identification and separation of different sound sources. The existing literature on the effect of duration on timbre recognition shows high performance for remarkably short time window lengths, but does not address the possible ef- fect of musical training. In this study, we applied an adap- tive procedure to investigate the effect of musical training on individual thresholds for instrument identification. A timbre discrimination task consisting of a 4-alternative for- ced choice (4AFC) of brief instrument sounds with varying duration was assigned to 16 test subjects using an adaptive staircase method. The effect of musical training has been investigated by dividing the participants into two groups: musicians and non-musicians. The experiment showed lo- west thresholds for the guitar sound and highest for the violin sound, with a high overall performance level, but no significant difference between the two groups. It is sug- gested that the test subjects adjust the weightings of the perceptual dimensions of timbre according to different de- grees of acoustic degradation of the stimuli, which are eva- luated both by plotting extracted audio features in a fea- ture space and by considering the timbral specificities of the four instruments.
Original languageEnglish
Title of host publicationProceedings of the 19th International Society for Music Information Retrieval Conference
Publisherdblp Computer Science Bibliography
Publication date2018
ISBN (Print)978-2-9540351-2-3
Publication statusPublished - 2018
Event19th International Society for Music Information Retrieval Conference, ISMIR 2018 - Paris, France
Duration: 23 Sep 201827 Sep 2018


Conference19th International Society for Music Information Retrieval Conference, ISMIR 2018

Fingerprint Dive into the research topics of 'Timbre Discrimination for Brief Instrument Sounds'. Together they form a unique fingerprint.

Cite this