Twenty-three male subjects each read a paragraph aloud and graphic level recordings of their readings were measured in order to determine the average sentence rate, phrase rate, and duration of phrase and sentence pauses. It was found that noninitial sentences and phrases were significantly faster in rate and that pauses between sentences were significantly longer in duration. The findings are consistent with a reduction of uncertainty hypothesis that rate increases and pause decreases as uncertainty decreases.

No abstract

Now, we are interested in finding that value of 0 which is closest to zero. This corresponds to maximizing cos0 in (A8) and also corresponds to minimizing the positive quantity G(u,v)= (logv-logu)/(v-u) (All) with respect to u and v over the triangular region in the u, v plane defined by (A10). Differentiating (All) with respect to u and v givesA12) and oO/Ou=El-v/u+log(v/u)•Ev-u• -2. (A13) These are negative quantities for the ranges of u and v given by (A10) or for 1/(2p-+-1) •< u/v •< (2p--1)/(2p+1). (A14)Consequently, G will be minimized if we take the largest values of u and v, namely, v= 2p+ 1 and u= v-2= 2p-1.These values correspond to qs=p and q•=p-1. From (AS) and (A15) we therefore have 0=arc cos[(2p--1)/(2p+ 1)-]:-. (A16) For a given value of p, we may conclude that an increase in the order of the lower ordered receiver will give an increase in signal-noise improvement for all 0 in the range 101 ,.

Generally quality defects of voice are presented as “independent of phonemes,” i.e., the sounds being articulated, and as the result of the unique wave composition of voice defective speakers. The purpose of this study was to investigate a voice quality disorder while varying specific speech sounds. The subjects with hoarse voice quality were selected to speak syllables in which the phonetic elements were controlled. The utterances were recorded and played to trained judges who rated the degree of hoarseness present in the utterance. Statistical analysis (analysis of variance) of the data collected indicated that significant differences in voice quality do exist as some function of the vowel and consonant content of the syllable. Further analysis (t test) reveals that high vowels result in significantly less judged voice quality defectiveness than the low vowels. Front-back vowel comparisons were generally less indicative of real differences. Of the consonants, it is seen that the voiceless, or whispered, plosives, such as t, p, and k, contribute to a judgment of less defectiveness than the other consonants. Generally, the findings are that voiceless consonants are judged less defective than voiced consonants, fricatives increase the judgment of hoarseness, the manner of articulation of consonants is related to perceived voice quality and place of articulation appears to have only slight correlation. In general the results suggest that the degree of perceived voice quality may be related in part to the secondary acoustic characteristics of syllables, such as duration, frequency, and intensity. Such investigations are now in progress.

No abstract