A speech corpus for multitalker communications research

Bolia, Robert S.; Nelson, W. Todd; Ericson, Mark A.; Simpson, Brian D.

doi:10.1121/1.428288

Cited by 462 publications

(356 citation statements)

References 4 publications

(6 reference statements)

Supporting

Mentioning

352

Contrasting

Order By: Relevance

“…The speech stimuli were sentences from the coordinate response measure (CRM) corpus (Bolia, Nelson, Ericson, & Simpson, 2000) with the structure "Ready [call sign] go to [color] [number] now," with eight call signs, four colors (white, red, green, and blue), and eight numbers (1)(2)(3)(4)(5)(6)(7)(8). The speech was processed to restrict the frequency content, as described in Arbogast et al (2002).…”

Section: Experiments 1 Methodsmentioning

confidence: 99%

Task-dependent costs in processing two simultaneous auditory stimuli

Gallun

Mason

Kidd

2007

Perception & Psychophysics

View full text Add to dashboard Cite

When presented with two speech utterances, one to each ear (i.e., dichotically), listeners can be asked to either (1) ignore one and report the other (selective attention) or (2) report both (divided attention). Despite more than 50 years of research on this topic, it is still not fully understood why different stimulus configurations exert such a great influence on the degree of success listeners experience when asked to either select one utterance or divide their auditory attention between two utterances. The goal of this study was to explore the theoretical distinction between selective and divided attention by presenting a method for distinguishing between the two processes in a single study. The new set of experimental results demonstrates the value of measuring both processes within the same study and provides an example of the sorts of hypotheses that such a technique can be used to generate.Two different types of experimental approaches have generally been used to examine the processing of multiple speech stimuli. The first type is what we will refer to as the dual-ear experiment (although, for comparisons, most such experiments have contained single-ear and/or diotic conditions as well). Dual-ear experiments are distinguished from other experiments involving speech stimuli by the fact that they include the presentation of one speech utterance to one ear and another utterance to the other ear, and in terms of results, they tend to emphasize the fact that under some conditions listeners can reliably report the information presented to the "target" ear with very few errors (e.g., Broadbent, 1958;Cherry, 1953;Moray, 1970;Treisman, 1964Treisman, , 1969Wood & Cowan, 1995).The second type is the informational masking approach, in which researchers have generally presented multiple speech stimuli to the same ear (although various types of dichotic presentation have also been used) and have emphasized the factors that lead to errors in processing only one of two simultaneously presented stimuli (e.g., Brungart, Simpson, Ericson, & Scott, 2001). The term informational masking is often defined in contrast to energetic masking, which refers to a reduction in performance that can be accounted for by the degree to which the masker overlaps the target at a set of peripheral analyzers (usually the cochlea or the auditory nerve). Informational masking, it is argued, is caused by nonperipheral factors, principally masker uncertainty and target-masker similarity (for examples, see Durlach et al., 2003;Neff, 1995;Neff & Green, 1987;Watson & Kelly, 1981; for a review, see Kidd, Mason, Richards, Gallun, & Durlach, 2006). A substantial number of studies have demonstrated that speech stimuli can interfere with each other in ways that are not easily captured by energetic masking alone (e.g., Arbogast, Mason, & Kidd, 2002;Brungart & Simpson, 2002;Brungart, Simpson, Darwin, Arbogast, & Kidd, 2005;Brungart et al., 2001;Freyman, Balakrishnan, & Helfer, 2001Freyman, Helfer, McCall, & Clifton, 1999;Kidd, Arbogast, Mason, & Gallun, 20...

show abstract

Section: Experiments 1 Methodsmentioning

confidence: 99%

Task-dependent costs in processing two simultaneous auditory stimuli

Gallun

Mason

Kidd

2007

Perception & Psychophysics

View full text Add to dashboard Cite

show abstract

“…Other factors that may influence intelligibility of speech synthesized by the ideal (or estimated) binary mask include the choice of the local SNR threshold(s), the masker type (modulated vs. steady-state), speech materials and input global SNR level. The effect of the local SNR threshold on speech intelligibility (and to some extent the masker type) was assessed in the study by Brungart et al (2006) using the Coordinate Response Measure (CRM) corpus (Bolia et al, 2000) as test material. The CRM test, however, is a closed-set test which is restricted to four phonetically distinct color alternatives and eight phonetically distinct number alternatives, both of which are easy to understand even in extremely noisy environments (Brungart et al, 2001a(Brungart et al, , 2001b.…”

Section: Introductionmentioning

confidence: 99%

Factors influencing intelligibility of ideal binary-masked speech: Implications for noise reduction

Loizou

2008

The Journal of the Acoustical Society of America

220

195

View full text Add to dashboard Cite

The application of the ideal binary mask to an auditory mixture has been shown to yield substantial improvements in intelligibility. This mask is commonly applied to the time-frequency (T-F) representation of a mixture signal and eliminates portions of a signal below a signal-to-noise-ratio (SNR) threshold while allowing others to pass through intact. The factors influencing intelligibility of ideal binary-masked speech are not well understood and are examined in the present study. Specifically, the effects of the local SNR threshold, input SNR level, masker type and errors introduced in estimating the ideal mask are examined. Consistent with previous studies, intelligibility of binary-masked stimuli is quite high even at -10 dB SNR for all maskers tested. Performance was affected the most when the masker dominated T-F units were wrongly labeled as target-dominated T-F units. Performance plateaued near 100% correct for SNR thresholds ranging from -20 dB to 5 dB. We believe the existence of the plateau region suggests that it is the pattern of the ideal binary mask that matters the most rather than the local SNR of each T-F unit. This pattern directs the listener's attention to where the target is and enables them to segregate speech effectively in multi-talker environments.

show abstract

“…The adaptive CCRM was based on the test developed by Bolia et al (2000) and Brungart (2001). It was used because of low contextual cues, and ease of task.…”

Section: Children's Coordinate Response Measure (Ccrm)mentioning

confidence: 99%