Measures Of Speech Production Research Articles

スピーチプロダクションの測定方法 : 正確さ、流暢さ、複雑さ

スピーチプロダクションの測定方法 : 正確さ、流暢さ、複雑さ

Tolerable hearing aid delays. III. Effects on speech production and perception of across-frequency variation in delay.

Signal processing strategies that attempt to mimic the frequency resolution of the healthy cochlea require finer frequency resolution at low frequencies than at high. The filtering required will cause more delay to the low-frequency end of the spectrum than to the high-frequency end, which may have disturbing effects. In a real-time application, making the delay constant across frequency would require the higher frequency components to be subjected to a compensating delay, leading to a larger overall delay, which may also give rise to disturbing effects (Stone & Moore, 1999, 2002). This experiment assessed the effects of frequency-dependent delay on subjective and objective measures of speech production and perception. Digital filtering was used to introduce across-frequency delays of 0, 4, 9, 15, or 24 msec. The low frequencies were always more delayed than the high frequencies, with a smooth variation in delay between the low and the high frequencies. In addition, there was a 2.5 msec delay that did not vary with frequency arising from the processes of analog-digital-analog conversion and equalization. Ten subjects with symmetric, bilateral, moderate hearing impairment of cochlear origin were fitted binaurally with Phonak PicoNet2 BTE linear hearing aids. The aids were programmed to provide insertion gains closely approximating the NAL (RP) (Byrne & Dillon, 1986) prescription for each subject and ear. The output of an omnidirectional microphone, attached on top of each aid, fed a digital filtering unit that introduced the across-frequency delay. Audio shoes reintroduced the signal back into the BTE aids. Subjects were tested in a counterbalanced order for their identification of vowel-consonant-vowel (VCV) syllables. In a second session, the same subjects spoke from a script: their utterances were recorded on tape, speech production rates were measured and subjective ratings of the disturbance of the delay were obtained. Subjects required some training to recognize the effects of the delay and rate them consistently. Subjective disturbance increased progressively with increasing across-frequency delay. A 9-msec delay was significantly more disturbing than a 0-msec delay. A delay of about 20 msec led to a mean rating of "disturbing." VCV identification decreased significantly once the across-frequency delay was 15 msec or greater. However, word production rates were not significantly affected by across-frequency delay over the range tested. Relatively small across-frequency delays (9 to 15 msec) give rise to significant changes in speech identification and subjective disturbance. A delay of 9 to 15 msec, constant across frequency, would have a smaller effect (Stone & Moore, 1999). It appears desirable to compensate, at least in part, for across-frequency delays introduced by any processing.

The relationship between early speech and later speech and language performance for children with cleft lip and palate

This study examined the relationship between speech measures at presurgery/9 months and postsurgery/13 months and speech and language performance at 21 months for children with cleft lip and palate and their noncleft peers. Comparisons were also made between the speech and lexical development of children with cleft lip and palate and noncleft children at 21 months of age. The participants included 30 children; 15 with cleft lip and palate and 15 noncleft children. Results revealed differences between the groups for several measures of speech and lexical development at 21 months. For the children with cleft palate, correlational analyses suggested that true stop production, both immediately before and after palatal surgery, was positively correlated with a majority of the speech production measures at 21 months. At postsurgery/13 months, true stop production was related to later vocabulary development, and size of true consonant inventory was related to all measure of speech production and one measure of lexical development at 21 months. For the noncleft group, true canonical babbling ratio at 13 months was the only measure that was significantly correlated with any of the speech and/or language measures at 21 months. The impact of clefting on prelinguistic and later speech and language skills is discussed.

Developmental change in variability of lip muscle activity during speech.

Compared to adults, children's speech production measures sometimes show higher trial-to-trial variability in both kinematic and acoustic analyses. A reasonable hypothesis is that this variability reflects variations in neural drive to muscles as the developing system explores different solutions to achieving vocal tract goals. We investigated that hypothesis in the present study by analyzing EMG waveforms produced across repetitions of a phrase spoken by 7-year-olds, 12-year-olds, and young adults. The EMG waveforms recorded via surface electrodes at upper lip sites were clearly modulated in a consistent manner corresponding to lip closure for the bilabial consonants in the utterance. Thus we were able to analyze the amplitude envelope of the rectified EMG with a phrase-level variability index previously used with kinematic data. Both the 7- and 12-year-old children were significantly more variable on repeated productions than the young adults. These results support the idea that children are using varying combinations of muscle activity to achieve phonetic goals. Even at age 12 years, these children were not adult-like in their performance. These and earlier kinematic studies of the oral motor system suggest that children retain their flexibility, employing more degrees of freedom than adults, to dynamically control lip aperture during speech. This strategy is adaptive given the many neurophysiological and biomechanical changes that occur during the transition from adolescence to adulthood.

Tolerable hearing aid delays. II. Estimation of limits imposed during speech production.

We used real-time processing in a wearable digital hearing aid to examine the effect of processing delay on normal-hearing participants while speaking. Objective and subjective data were recorded so as to permit analysis of both the production and perception of speech read aloud from a script. We also asked participants to rate the disturbance of the echo introduced by the delay. Thirty-two (16M, 16F) participants were fitted binaurally with behind-the-ear (BTE) aids connected to a digital processor. A 4 mm Libby horn, surrounded by an expanding foam earplug, conducted processed sound into each ear canal. The processor provided either linear processing or three-channel, fast-acting wide dynamic range compression, independently to each ear. Insertion gains were set, using a KEMAR manikin, to be 0 dB over a wide frequency range, for frontally presented speech with a free field level of 65 dB SPL. Additionally, the aids introduced one of four selectable delays (7 to 43 msec) between the BTE microphone and receiver. After a short period of acclimatization, each participant read 16 prose passages of about 500 words in length in each of two similar-sized rooms with markedly different acoustics: reverberant and nonreverberant. For each passage, a subjective rating of the level of disturbance of the perceived echo was recorded, as well as simultaneous recordings from a microphone and a Laryngograph, which directly records glottal pulses. Disturbance ratings generally increased monotonically with increasing delay. Averaged results show that a delay between 25 and 30 msec is rated as "disturbing." Measures were also taken of word production rate, speech level and range of level as well as fundamental frequency and range of fundamental frequency. For these measures of speech production, there was no significant effect until the delay exceeded 30 msec. There was little effect of acoustic environment or aid processing (linear or compression). The acceptability of delays introduced by digital hearing aids is primarily determined by aspects of the perception of self-generated speech. Speech production, on average, is hardly affected unless the processing delay exceeds 30 msec. The permissible limit of 20 to 30 msec is smaller than the delays at which audio-visual integration is disrupted.

Harry Levitt legacies: Studies of speech perception and intelligibility in children with cochlear implants

Harry Levitt has contributed a substantial body of knowledge regarding the speech perception and production skills of hearing-impaired children. In this paper, Dr. Levitt will be honored by describing the speech perception, speech intelligibility, and acoustic characteristics of 136 cochlear implanted children between the ages of 8 and 9 who have at least 4 years experience with a Nucleus multichannel cochlear implant. Half of the children used auditory-oral modes of communication and the other half used total communication. The questions addressed are as follows. (a) Do children implanted before 5 years of age develop intelligible speech? (b) Are speech perception skills reflected in the acoustic measurements of speech in children with cochlear implants? (c) What factors are associated with intelligible speech in children with cochlear implants? The presentation will describe the population demographics, speech perception measures (including open and closed set speech perception performance), and measures of speech production (including speech intelligibility, communication breakdowns, parental rating scales, and acoustic characteristics). Factors contributing to high levels of speech perception and intelligibility will be described and contrasted. Implications for the development of intelligible speech will be discussed, particularly as it relates to mode of communication. [Work supported by NIH-NIDCD.]

Spectral characteristics of /s/ and /∫/: Rate effects in Parkinson’s disease

Comtemporary speech production theory proposes that speakers adjust rate to enhance or maximize acoustic-perceptual distinctiveness for listeners. In fact, the suggestion that rate impacts acoustic-perceptual distinctiveness is one rationale for using rate manipulation as a therapy to improve speech intelligibility in dysarthria. The relationship between rate manipulation, acoustic measures of production, and perceptual impressions of speech in dysarthria is poorly understood, however. For example, some studies report rate-induced acoustic changes in speech that are systematically tied to perceptual impressions. Other studies report no effect of rate on speech production characteristics in dysarthria. Further study of the relationship between speech production measures and rate variation in dysarthria therefore appears warranted. The present study examined spectral characteristics of word-initial /s/ and /∫/ produced at a variety of rates by speakers with hypokinetic dysarthria associated with Parkinson’s disease. Healthy speakers were studied for comparison purposes. Fricatives were produced in the words shores, shies, size, and soys embedded in a carrier phrase. A graded rate task was used to elicit an extensive range of rates. Spectral moment coefficients were used to characterize fricatives; phrase duration was used to index rate. The relationship between rate variation and spectral moment coefficients was quantified via regression analysis.

Speech production and language development in pediatric cochlear implant users.

Cochlear implantation has been very successful in postlingually deafened children. Implants provide them with substantial levels of speech reception, which in turn allows them to achieve better intelligibility, speech production, and English language proficiency. Congenitally or prelingually deafened children, while less successful than the aforementioned group, have also shown benefit in speech production and language measures. Performance varies greatly in this group, but there appears to be a trend towards better results when implantation is performed earlier in life. This presentation will analyze measures of speech production and language ability in children deafened before age 3 (who comprise the vast majority of pediatric cochlear implant candidates) as well as in children who are later deafened. Individual and group data will be discussed in view of the subjects’ communication mode (oral or total communication), age at deafening, age at implant, and years of experience with the implant. [Work supported by NIDCD Grant Nos. DC00064 and DC00423.]

Results for two children using a multiple-electrode intracochlear implant.

Assessments in speech perception, speech production, and language skills were conducted on two children, 5 and 10 years old at the time of surgery, using the Nucleus multiple-electrode cochlear implant. Data were collected pre- and post-operatively to measure changes in performance over time. For closed-set speech perception tests in the audition alone condition, post-operative performance was generally better than pre-operative performance and performance improved post-operatively for both patients. In closed-set vowel and consonant identification and open-set sentence perception for the older patient, post-operative improvements from the vision alone to the auditory-visual condition were recorded and performance improved post-operatively in both conditions. In all measures of speech production for both patients, post-operative scores were higher than pre-operative scores and performance improved post-operatively. In language skills, post-operative scores were higher than pre-operative scores and scores improved post-operatively in all measures for both patients.

Palatography: A comparison between custom‐made and “flexible” artificial palates for speech production measures

While dynamic palatography, a method of measuring tongue‐palate contact over time, has a long history, the method has seen limited use because of the need to custom‐make a palate for each subject. Recently, the Rion Company has developed a “universal” semiflexible palate, with 64 contact points, in several sizes. This methodological study was designed to assess the comparability of data from flexible and custom‐made palates, and the stability and reliability of flexible palate measures. Three speakers repeated multiple tokens of /isi/ and /asa/ with both palate types, and with flexible palates removed and replaced. While test‐retest reliability was satisfactory for flexible palates, the custom‐flexible comparisons revealed the need to develop measures of subject‐to‐subject differences in palate size and shape, if the technique is to develop its potential for large‐scale studies of articulation in normal and clinical populations. Comments will be made on analysis software, sampling rate, and contact‐point distribution. [Work supported by NIH Grant NS‐13617.]

The effect of wearing aerodynamic measuring devices on speech articulation

Aerodynamic measures of speech production have been used to obtain insights into the dynamic properties of the vocal tract during speech production. It is unclear, however, how wearing devices such as the pneumotachograph and oral pressure transducers affects the articulatory processes being studied. The present research examined the articulatory characteristic of speech production by two normal speakers under two conditions: (1) while wearing a Rothenberg type pneumotachograph mask and oral pressure transducer and (2) without these devices. The x-ray microbeam data were obtained for lip, jaw, and tongue movement during multiple productions of voiced and unvoiced plosives in varied vowel contexts. Preliminary analysis suggests that articulatory dynamics appear not to change as a result of wearing the mask, although movement amplitudes may. Comparisons between the two conditions will be presented, together with implications for use of such measures as indices of speech articulatory behavior.

Efficacy of a distinctive feature model of therapy for hearing-impaired adolescents.

Extensive measures of speech production and perception were secured before and after the individualized training of 75 hearing-impaired postsecondary students to evaluate the efficacy of therapy. Additional measures of hearing, reading, writing, nonverbal intelligence, and manual and simultaneous communication reception were made to explore variables that might influence response to training. Significant improvements in vowel, word and sentence production, and word and sentence perception were observed. Pretherapy measures of speech production and perception were the most powerful factors explaining variability in posttest scores. Other variables predicting the outcome of speech therapy were residual hearing and linguistic competence. Clinical implications are discussed.

Measures of speech production

The author discusses the mechanisms of speech production, focusing on the measurement of its motor aspects. He examines the multiple representation levels at which physiological events that underlie speech production can be investigated, and examines measurement of neuromuscular activity, structure movement (e.g. kinematics and dynamics), aerodynamic phenomena, and/or the acoustic output. He describes two general approaches. Studies can focus on one system and obtain data at multiple levels; for example, simultaneous recording of EMG (electromyogram) and/or kinematic signals, as well as the acoustic signal; this permits comprehensive assessment of physiologic events within a particular system. Alternatively, studies can obtain recordings of the acoustic signal simultaneously with EMG and/or kinematic signals taken from several systems; this permits assessment of the organization of physiologic events within, as well as between, systems. The author illustrates these methods in a study of stuttering.

Age‐sex variations in maximum performance measures of speech production

The maximum performance measures of speech production include: phonation volume, maximum phonation time, maximum expiratory pressure, maximum vocal SPL, physiological vocal frequency range, maximum articulatory force, and maximum syllable repetition rate. A survey of the normative data seemed in order, given the frequent clinical use of these measures and the availability of a large, although scattered, archival database for normal speakers. This report reviews published data for each measure and considers the adequacy of these data for clinical application. The data are found to be deficient in several respects: (1) the effects of subject age and sex are incompletely documented; (2) test procedures are not standardized; (3) the validity of a maximum performance estimate is open to question for most, if not all, measures; (4) the variability of the data within and across studies is very large; and (5) a theoretical framework for the interpretation of the measures is either lacking or deficient. Recommenda...

Evaluation of speech production of the hearing impaired: some benefits of forced-choice testing.

The speech production performance of 16 hearing-impaired subjects (thresholds 82-110 dB HL) was measured using both experienced and inexperienced listeners. Four types of measure were obtained: score on a forced-choice test requiring the production of eight segmental contrasts, recognition probability for phonemes produced in consonant-vowel-consonant (CVC) words, recognition probability for CVC words produced in a simple carrier phrase, and recognition probability for words produced in sentence context. The experienced listeners gave higher scores than did the inexperienced listeners, but the magnitude of the effect decreased with decreasing linguistic redundancy in the test material, falling to only 1.3 percentage points for the forced-choice test. The pattern of performance across the eight contrasts of the forced-choice test changed significantly from subject to subject. Performance on the contrast test accounted for 70% of the variance in the scores on the sentence test. The results demonstrate that forced-choice tests can be used to provide speech production measures that are virtually independent of listener experience, that provide analytic detail about an individual's production of phonetic contrasts, and that are reasonably predictive of the intelligibility of speech produced in more natural communicative settings.

Speech-production measures of speech perception: rapid shadowing of VCV syllables.

Five listeners rapidly repeated ("shadowed") a random presentation of the vowel-consonant vowels (VCV's)/aba, apa, ama, aka, aga/. Initial vowel duration was varied to eliminate it as a temporal cue to the occurrence of the consonant. These shadowing, choice reaction times (RT's) were compared to simple RT's obtained when listeners always produced /aba/ or /ba/ to the same syllables. Both /aba/ and shadowing reactions were extremely fast (170 to 240 ms). Latency differences between the two tasks were attributable to differences in the point at which cues sufficient for responding were present. These results suggest that speech-perception decisions in shadowing are directly available to, and are perhaps made to occur at a point comparable to the consonantal release seen for the simple /aba/ responses. This result suggests that the motor organization required for a /ba/ response includes an implicit time interval appropriate for a consonantal closure.