V. Sivonen, M. Alanko, H. Gamper, H. Kari, O. Rummukainen, and V. Pulkki
Binaural hearing is beneficial in suppressing the audibility of echoes in reverberant environments. Echo suppression has typically been investigated utilising a single reflection, and often at high direct-to- reverberant energy ratios (DRRs) where the reflection is barely audible. In natural environments, however, reverberation comprises a large number of reflections and DRR may be relatively low. The aim of this investigation was to quantify binaural echo suppression in the presence of reverberation by assessing its salience in monaural and binaural listening. A short speech segment and an impulsive finger snap were used as direct sound, and a simple reverberation model with frequency-dependent T60 between 0.5 and 2.0 s was used to create stimuli with different DRRs. The stimuli were spatialised using non-individual head-related transfer functions and played back over headphones. In an adaptive matching paradigm, a diotic or dichotic reference sound was fixed at 0 dB DRR, and the DRR of a monaural or diotic comparison sound was varied via a two-alternative forced-choice procedure. Twelve listeners participated, and on each trial, their task was to judge which of the two sounds, the reference or the comparison played back in random order, was more reverberant. The results show that between 2-6 dB higher DRR is needed for a monaural sound to be matched in perceived reverberation with the binaural sound, the effect being largest for the impulsive sound and when matched against the dichotic reference. This is in agreement with earlier studies on echo suppression of a single reflection.
|Published in||Proc. Forum Acusticum|