Javascript must be enabled for the correct page display

Classification of signal components in cochleograms

Oerle, P. van (2006) Classification of signal components in cochleograms. Master's Thesis / Essay, Artificial Intelligence.

AI_Ma_2006_PvanOerle.CV.pdf - Published Version

Download (2MB) | Preview


Distinguishing between the sounds in our environment is quite trivial for humans. We can easily distinguish between sounds that originated from different sources, and routinely isolate single sources. Since the '50s, the perceptual segregation of sound into components has been object of study, but a definitive method to accomplish this, seemingly so very simple, feat has not been found. To facilitate the analysis of sound - to accomplish, for example, speech recognition or instrument separation - some form of base separation wifi have to be performed. The basic components of sound can be broadly separated into three categories: pulses, sines, and noise-like components. Detecting these base components of sound would allow for later application of algorithms to combine them into greater, more complicated components defining the structure of sound. A method to accomplish the detection of such base components was investigated, using Sound Intelligence's Cochlear Analysis system and existing algorithms written by Tjeerd Andringa and Dirkjan Krijnders. A simple system was made which tries to detect components in cochleograms, and resynthesises a signal based on the detected components. An analysis of the performance of the system was made by assessing the results of the resynthesis of several testsignals.

Item Type: Thesis (Master's Thesis / Essay)
Degree programme: Artificial Intelligence
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 15 Feb 2018 07:30
Last Modified: 15 Feb 2018 07:30

Actions (login required)

View Item View Item