These experiments investigate how the identification of vowel-like harmonic complexes, similar to those used by Leek et al. [J. Acoust. Soc. Am. 81, 148–154 (1987)], is affected by spectral contrast, overall level, component phase, and fundamental frequency (F0). Four normally hearing subjects were required to identify which of six vowel-like harmonic complexes was presented on each trial. The test stimuli were complex tones containing the first 35 harmonics of a 100-Hz fundamental or the first 70 harmonics of a 50-Hz fundamental. All of the harmonics with frequencies below 3000 Hz were equal in amplitude except for three pairs of successive harmonics which were located at the first, second, and third formant frequency values, and incremented in level by 1, 2, 4, 8, and 16 dB relative to the other components. Three overall levels were used, 85, 65, and 45 dB SPL, and harmonics were added in either cosine or random phase. The results indicated that identification was better for cosine phase than for random phase, except for the 100-Hz fundamental at 45 dB SPL. The difference between the two phase conditions increased with increasing presentation level and with decreasing fundamental frequency. The results are explained in terms of the waveforms that would occur at the outputs of different auditory filters. It does not appear necessary to invoke nonlinear enhancement mechanisms to explain the results, although an influence of such mechanisms cannot be ruled out.
展开▼
机译：这些实验研究了类似于Leek等人的元音类谐波复合物的识别方法。 [J. co Soc。上午。 81，148–154（1987）]，受光谱对比度，总体水平，分量相位和基频（F0）的影响。每个实验需要四个正常听力的受试者来识别六个元音样谐波复合物中的哪个。测试刺激是复杂的音调，包含100 Hz基本频率的前35个谐波或50 Hz基本频率的前70个谐波。除了位于第一，第二和第三共振峰频率值上的三对连续谐波，并且其电平分别增加1、2、4、8和16外，所有频率低于3000 Hz的谐波的幅度均相等。相对于其他分量的dB。使用了三个整体级别，分别为85、65和45 dB SPL，并且在余弦或随机相位中添加了谐波。结果表明，除了在45 dB SPL时的100 Hz基本频率以外，余弦相位的识别性能优于随机相位的识别功能。两个相位条件之间的差异随着呈现水平的提高和基本频率的降低而增大。根据将在不同听觉滤波器的输出处出现的波形来解释结果。尽管不能排除这种机制的影响，但似乎没有必要调用非线性增强机制来解释结果。
展开▼
