Auditorium speech quality : effect of delayed same speech maskers

Abstract

This study investigates the masking effect of a simulated delayed same-signal reflection on the direct sound.  The main objective is to contribute to understanding the phenomenon of speech masked by a delayed same-speech reflection.

Three experiments were designed.  In the first experiment, the results show that the masking thresholds obtained are affected by the level difference between the signal and the reflection, the time difference between the signal and the reflection, and the direction of the masker relative to the direct sound.  The second experiment suggests that the speech signal can be detected in the silence gap as well as in the low amplitude consonant part of the speech.  In addition, it also suggests that low frequencies, such as vowels, may not always mask high frequencies as consonants in the context of overlap masking within the range of parameters tested.  In the last experiment, the direction of reflection and the effects of head rotation affect the masking threshold.

Following the three experiments, it can be concluded that the sound quality in any room may be assumed to be good when the delay is short.  Unfortunately, this statement cannot be applied when sound reinforcement system is used.  Interestingly, when the sound reinforcement system with multi-loudspeakers is used, the short  delay may not always provide the good sound quality.  For further development of objective parameter determining speech quality based on masking effect in a room, the direction of the reflection is significant.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata