Lombard Speech Detection in Case of Spatial Separation between Noise Source and Speakers of Different Genders

The spatial selectivity of hearing to speech signals was studiedwhen the target signal and interference were separated by distancebetween their sources and the listener. In the work, the hypothesisabout the improvement of hearing selectivity due to more intensiveactivation of the high-frequency binaural mechanism due to the shiftof the speaker’s voice spectrum occurs in noisy environment towardshigh frequencies, was tested. The thresholds for detecting the targetsignal—a two-syllable word uttered by male or female, were evaluatedin the two-alternative two-interval forced-choice paradigm in 4series. Series differed by the type of target signal (normal orLombard speech) and the location of target source and noise (multi-talkernoise) one. The both sources were located at a distance of 1 and4 m opposite the subject at the level of his head. The detectionthreshold was defined as the ratio of signal and noise levels atthe listener's place (S/N). The threshold for detecting the targetsignal (male and female speaker voices together) was –11 dB S/Nfor normal as well as Lombard speech. It did not depend on whichof the sources—the target signal or noise, was closer to the listener.In normal speech, the detection thresholds on average differed formale and female voices, but the difference was not significant.In Lombard speech, these thresholds were significantly different:for a male voice, the threshold at a detection level of 0.67 was–10 dB S/N, and for a female voice was –12.5 dB S/N.