Comodulation masking release in consonant recognition

Abstract

Comodulation masking release (CMR) refers to an improvement in the detection threshold of a signal masked by noise with coherent amplitude fluctuation across frequency, as compared to noise without the envelope coherence. The present study tested whether such an advantage for signal detection would facilitate the identification of speech phonemes. Consonant identification of bandpass speech was measured under the following three masker conditions: (1) a single band of noise in the speech band ("on-frequency" masker); (2) two bands of noise, one in the on-frequency band and the other in the "flanking band," with coherence of temporal envelope fluctuation between the two bands (comodulation); and (3) two bands of noise (on-frequency band and flanking band), without the coherence of the envelopes (noncomodulation). A pilot experiment with a small number of consonant tokens was followed by the main experiment with 12 consonants and the following masking conditions: three frequency locations of the flanking band and two masker levels. Results showed that in all conditions, the comodulation condition provided higher identification scores than the noncomodulation condition, and the difference in score was 3.5% on average. No significant difference was observed between the on-frequency only condition and the comodulation condition, i.e., an "unmasking" effect by the addition of a comodulated flaking band was not observed. The positive effect of CMR on consonant recognition found in the present study endorses a "cued-listening" theory, rather than an envelope correlation theory, as a basis of CMR in a suprathreshold task.

Comodulation masking release in speech identification with real and simulated cochlear-implant hearing

“…Previous studies of CMR in speech recognition have used band-pass filtered speech presented to NH listeners (Grose and Hall, 1992;Kwon, 2002;Buss et al, 2003). In one previous study, when stimuli consisted of either bandpass filtered words with six possible forced-choice responses in each trial or open-set sentence material, no CMR was observed (Grose and Hall, 1992).…”

“…In one previous study, when stimuli consisted of either bandpass filtered words with six possible forced-choice responses in each trial or open-set sentence material, no CMR was observed (Grose and Hall, 1992). In another study, where listeners were asked to identify intervocalic consonant syllables and performance was measured in a range between 40% and 65% of overall percent correct performance levels, CMR caused less than 4% of improvement in performance (Kwon, 2002). In a third study, when target speech material consisted of spondees with either two or four forced-choice alternatives, CMR was about 3 dB for a two-alternative task or 1 dB for a four-alternative task (Buss et al, 2003).…”

“…At least for NH listeners, this "comodulation masking release (CMR)" has also been demonstrated for speech tasks with noise interference. For NH listeners performing tasks with narrowband filtered speech in noise, CMR can improve speech detection thresholds by 11 dB and speech discrimination thresholds by 6-7 dB in a two-alternative forced choice vowel discrimination task (Grose and Hall, 1992) but is modestly sized or absent for consonant identification (Grose and Hall, 1992;Kwon, 2002) and open set speech identification (Grose and Hall, 1992;Festen, 1993;Buss et al, 2003). Moreover, vocode simulations of CI processing suggest that CMR may improve tone detection in CI listeners (Pierzycki and Seeber, 2010).…”

Asynchronous glimpsing of speech: Spread of masking and task set-size

“…In short, CMR is the improvement in detection thresholds seen when comodulated off-frequency maskers are added to an onfrequency masked target. While CMR could have played some role in the results of Howard-Jones and Rosen, it is unlikely to account for synchronous/asynchronous AM differences that were on the order of 20 dB; studies have shown CMR to have relatively small contributions to performance with supra-threshold stimuli, including speech (Grose and Hall, 1992;Hall et al, 1997;Kwon, 2002;Buss et al, 2003).…”

Effects of directional microphone and adaptive multichannel noise reduction algorithm on cochlear implant performance

“…The level of the speech spectrum calibration noise was adjusted to be 56 dB SPL from individual speakers which resulted in a uniform noise field with an overall level of 65 dB SPL. An uncorrelated noise field was used to avoid the comodulation masking release effect which is a result of correlated noise field and could yield an increase in speech understanding scores compared to an uncorrelated noise field ͑Cox and Bisset, 1984; Grose and Hall, 1992;Kwon, 2002;Moore, 1990͒. A continuous noise field, instead of the gated noise used in the original HINT test, was utilized to ensure that the noise reduction algorithm was always engaged.…”