Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.599069
Title: The influence of loudspeaker type on timbre perception
Author: Flanagan, S. A.
Awarding Body: University of Cambridge
Current Institution: University of Cambridge
Date of Award: 2005
Availability of Full Text:
Full text unavailable from EThOS. Please contact the current institution’s library for further details.
Abstract:
In any typical listening environment the sounds we hear are the combination of the direct sound from the source and its reflections. However, it is usually possible to identify sound sources with little difficulty. It has been suggested that a “timbral constancy” mechanisms is at work. Theories of how this timbral constancy may come about are discussed. One theory is that there is a ‘precedence-like’ effect for timbre. The first arriving direct sound is unaffected by reflections and carries the uncoloured timbral information. The second theory is that there is some form of perceptual compensation based on the long-term spectra reaching the ears. A series of experiments was carried out to test these two theories. This was facilitated by the use of two types of loudspeakers with very different acoustic radiation properties. The first was a traditional cone loudspeaker. The second was a distributed mode loudspeaker (DML). The first three experiments investigated discrimination of a “rippled” spectral profile from a “flat” profile. Thresholds were not significantly different for the two types of loudspeaker. A significant binaural advantage was observed for both types of loudspeakers and this is discussed in terms of a ‘central spectrum’ model. The third experiment measured identification of vowel-like sounds. Performance was above chance for a spectral contrast of 2 dB. Surprisingly, a significant effect of stimulus phase was observed; performance was better for cosine phase than for random phase. The fourth experiment investigated the discrimination of group delay in click-like sounds. For presentation via headphones or in a low-reverberation room, threshold were low (~2 ms) and independent of the centre frequency of the delayed part. In a reverberant room thresholds were higher and increased with decreasing centre frequency of the delay.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.599069  DOI: Not available
Share: