Differential contribution of envelope fluctuations across frequency to consonant identification in quiet

Abstract

Two experiments investigated the effects of critical bandwidth and frequency region on the use of temporal envelope cues for speech. In both experiments, spectral details were reduced using vocoder processing. In experiment 1, consonant identification scores were measured in a condition for which the cutoff frequency of the envelope extractor was half the critical bandwidth (HCB) of the auditory filters centered on each analysis band. Results showed that performance is similar to those obtained in conditions for which the envelope cutoff was set to 160 Hz or above. Experiment 2 evaluated the impact of setting the cutoff frequency of the envelope extractor to values of 4, 8, and 16 Hz or to HCB in one or two contiguous bands for an eight-band vocoder. The cutoff was set to 16 Hz for all the other bands. Overall, consonant identification was not affected by removing envelope fluctuations above 4 Hz in the low- and high-frequency bands. In contrast, speech intelligibility decreased as the cutoff frequency was decreased in the midfrequency region from 16 to 4 Hz. The behavioral results were fairly consistent with a physical analysis of the stimuli, suggesting that clearly measurable envelope fluctuations cannot be attenuated without affecting speech intelligibility.

Figure 1

Logarithm of the amplitude components of the modulation spectrum of a Gaussian noise, as computed at the output of a bandpass filter centered at 1000 Hz. The bandwidths of the bandpass filter were 1024 and 128 Hz for the upper and the lower panels, respectively. Prior to computation, the envelope of each band of noise was low pass filtered at 128 Hz.

Figure 2

Mean percent scores for consonant identification as a function of the number of bands. The parameter is the cutoff frequency of the envelope low-pass filter.

Figure 3

Mean percent scores for consonant identification as a function of the band tested (SING condition). The parameter is the cutoff frequency of the envelope low-pass filter used for the tested band. Dashed and solid lines show the mean percent scores when the envelope filter for all bands was set to the same cutoff frequency, either 16 Hz or half the critical bandwidth (HCB8), respectively.

Figure 4

Mean percent scores for consonant identification as a function of the bands tested (CONT condition). The parameter is the cutoff frequency of the envelope low-pass filter used for the tested bands. Dashed and solid lines show the mean percent scores when the envelope filter for all bands was set to the same cutoff frequency, either 16 Hz or half the critical bandwidth (HCB8), respectively.

Figure 5

The upper, middle, and lower panels show the percent correct relative to the base line condition (all bands at 16 Hz) in the 4 Hz, 8 Hz, and HCB conditions respectively, as a function of the tested bands. In each panel, the filled and unfilled circles show the results for the sum Sing and the CONT condition, respectively.

Figure 6

Logarithm of the amplitude components of the modulation spectrum computed for all 64 VCVs used in the experiments as a function of the analysis band. The same eight bandpass filters used in the HCB condition were used for the analysis (i.e., the envelope low-pass filter cutoff frequency was equal to half the bandwidth of an auditory filter). Levels of gray indicate a 6 dB amplitude range. For clarity, lines on the modulation frequency axis corresponding to an envelope cutoff condition are in bold (4, 8, and 16 Hz).