Temporal coherence effects on voice attribution in multi-speaker stream segregation.

Area of Science:

• Auditory perception
• Speech processing
• Psychoacoustics

Background:

• The principle of temporal coherence suggests that synchronized auditory streams are perceived as unified, while desynchronized streams are segregated.
• Previous research using simple stimuli demonstrated temporal coherence's role in auditory stream segregation.
• The applicability of these principles to complex auditory signals like human speech remains less understood.

Purpose of the Study:

• To investigate whether temporal coherence influences auditory stream segregation in complex speech.
• To test the prediction that temporally coherent speech forms a unified stream, while incoherent speech forms separate streams.

Main Methods:

• Twenty normal-hearing listeners participated in the study.
• Listeners identified the final word of a target talker's speech.
• Stimuli involved temporally coherent or incoherent speech phrases preceding the target, or silence.

Main Results:

• No significant differences in speech intelligibility or stream segregation performance were found between temporally coherent and incoherent speech conditions.
• Listener performance did not differ significantly when compared to a silence control condition.

Conclusions:

• The principle of temporal coherence may not generalize to complex auditory streams like human speech.
• Further research is needed to understand auditory stream segregation with more complex acoustic stimuli.