Neutral Vowel Research Articles

Computer recognition of the speech of adults with cerebral palsy and dysarthria

Ten speakers with cerebral palsy and dysarthria and 13 nondisabled speakers were tested using a Shadow VET/2 speech recognition system installed in an Apple Ile computer. The speakers each produced four types of stimuli: 12 consonants followed by a neutral vowel, 12 vowels in an h-d environment, 12 words defined as easy and 12 words defined as difficult for dysarthric speakers. Using all four types of stimuli, the dysarthric speakers had scores above chance, but they had significantly fewer stimuli recognized by the computer than the nondisabled speakers. For the dysarthric group, vowels in an h-d environment and both types of words were recognized significantly more often than the consonants followed by a neutral vowel. For the nondisabled group, the vowels in an h-d environment were recognized significantly more often than consonants followed by a neutral vowel, and both types of words were recognized more often than vowels. Confusion matrices and a discussion of item errors were used to further compare...

Stimulus order effects in vowel discrimination

In same-different discrimination tasks employing isolated vowel sounds, subjects often give significantly more "different" responses to one order of two stimuli than to the other order. Cowan and Morse [J. Acoust. Soc. Am. 79, 500-507 (1986)] proposed a neutralization hypothesis to account for such effects: The first vowel in a pair is assumed to change its quality in memory in the direction of the neutral vowel, schwa. Three experiments were conducted using a variety of vowels and some initial support for the hypothesis was obtained, using a large stimulus set, but conflicting evidence with smaller stimulus sets. Rather than becoming more similar to schwa, the first vowel in a pair seems to drift toward the interior of the stimulus range employed in a given test. Several possible explanations are discussed for this tendency and its relation to presentation order effects obtained in other psychophysical paradigms is noted.

Ultra-high rate digital sampling: Jitter, shimmer, and spectral perturbation of speech

The glottal function was obtained through the use of a reflectionless tube. This tube acted as a pseudoinfinite termination of the vocal tract. Male and female speakers phonated a neutral vowel while target matching to a 125- and 210-Hz tone, respectively. The glottal waveform was digitized at 1 000 samples per second with a resolution of 12 bits. Each cycle was partitioned at an interpolated zero-crossing. This procedure provided an accurate determination of the period of each glottal pulse. The spectral content of each individual cycle was determined by discrete Fourier analysis. The amplitude and phase of the first 20 harmonics were determined for each cycle. These harmonics were utilized to determine separate perturbation values, as well as an overall amplitude measure. The data showed a systematic change in the spectral character of the sustained vowel source signal. Spectral perturbation analysis may be a more inclusive measure of waveshape changes than jitter or shimmer.

Hungarian neutral vowels

This paper reports on recent empirical investigations of Hungarian which call into question several widely accepted claims about neutral vowels in Hungarian. Subjects treated e more like a front harmonic vowel than a neutral vowel, they were more likely to use back vowel suffixes with mixed vowel roots with e and é in the last syllable if the roots were disyllabic than if they were longer, and most subjects responded with front harmonic suffix vowels for most roots with neutral vowels in the final two syllables.

Cycle to cycle spectral perturbations in the voices of male and female speakers

The glottal function was obtained through the use of a reflectionless tube. This tube acted as a pseudoinfinite termination of the vocal tract. Male and female speakers phonated a neutral vowel while target matching to a 125- and a 210-Hz tone, respectively. The glottal waveform was digitized at 40k samples/s and each cycle was partitioned at zero crossings. This procedure provided an accurate determination of the period of each glottal pulse. Discrete Fourier analysis on each individual cycle indicated changes in the spectral content. The amplitude and phase of the first ten harmonics were utilized to obtain separate perturbation values as well as an overall amplitude measure. The data showed a systematic change in the spectral character of the sustained vowel source signal. Spectral perturbation analaysis may be a more inclusive measure of waveshape changes than jitter or shimmer.

Transparency in Hungarian vowel harmony

Phonologists have known for some time that the so-called ‘standard’ theory of generative phonology is not adequate for the analysis of vowel harmony. Ringen (1975, 1977, 1980) suggests that some of the problems can be solved by abandoning the assumption that phonological representations are fully specified. Clements (1977b, 1980) suggests that vowel harmony should be analysed autosegmentally. Underspecification theory, developed in the recent work of Kiparsky, Archangeli and Pulleyblank, incorporates both of these proposals. This paper considers how Hungarian can be analysed within this theory. It is shown that by adopting Goldsmith's (1985) proposal that vowel harmony in Hungarian involves the spreading of the feature [−back], the transparent (neutral) vowels in Hungarian are derived because the redundancy rule assigning [−back] to these vowels, although available, does not apply early in the derivation because its structural description is not met.

The contribution of the excitatory source to the perception of neutral vowels in stuttered speech

The vowel in part-word repetitions in stuttered speech often sounds neutralized. In the present article, measurements of the excitatory source made during such episodes of dysfluency are reported. These measurements show that, compared with fluent utterances, the glottal volume velocities are lower in amplitude and shorter in duration and that the energy occurs more towards the low-frequency end of the spectrum. In a first perceptual experiment, the effects of varying the amplitude and duration of the glottal source were assessed. The glottal volume velocity recordings of the /ae/ vowels used in the analyses were employed as driving sources for an articulatory synthesizer so that judgments about the vowel quality could be made. With dysfluent glottal sources (either as spoken or by editing a fluent source so that it was low in amplitude and brief), the vowels sounded more neutralized than with fluent glottal sources (as spoken or by editing a dysfluent source to increase its amplitude and lengthen it). In a second perceptual experiment, synthetic glottal volume velocities were used to verify these findings and to assess the influence of the low-frequency emphasis in the dysfluent speech. This experiment showed that spectral bias and duration both cause stuttered vowels to sound neutralized.

Cycle to cycle spectral perturbations in the voices of female speakers

The glottal function was obtained through the use of a reflectionless tube. This tube acted as a pseudoinfinite termination of the vocal tract. Ten female speakers phonated a neutral vowel while target matching to a 210‐Hz tone. The glottal waveform was digitized and each cycle was partitioned at zero crossings. This procedure provided an accurate determination of the period and separate timing measures for the positive and negative portions of the glottal pulse. Discrete Fourier analysis on each individual cycle indicated changes in the spectral content. The data showed a systematic change in the spectral parameters related to the open‐close ratio for each cycle. Spectral perturbation analysis may be a more inclusive measure of waveshape changes than jitter or shimmer.

Vowel harmony, neutral vowels and autosegmental theory

A revision of McCarthy's analysis of Montaiies vowel harmony is proposed in which autosegmental tiers constitute a separate dimension of the representation whose feature specifications can be given independently. Segments which do not undergo harmony (including neutral vowels) are specified on the segmental tier and are not affected when a word is associated to the autosegmental tier (cf. Booij (1984)). By allowing rules to make reference to the specification of a whole tier within a harmony domain it is possible to simplify McCarthy's treatment considerably, in particular dispensing with his ‘delinking’ rule and consequently his 'Duke of York' analysis. It is also not necessary to invoke extrinsic rule ordering.

Frequency and amplitude perturbations of the human voice determined with a reflectionless tube

In this study, the glottal function was obtained through the use of a reflectionless tube. This tube acted as a pseudoinfinite termination of the vocal tract. Male subjects phonated a neutral vowel (schwa) while target matching to tones throughout the vocal range. These waveforms were digitized at 61 440 samples per second for 1 s. Amplitude and frequency perturbations were obtained for cycle‐to‐cycle variations, as well as a normalized distribution. Spectral analysis of the digitized waveforms provided a simultaneous description of frequency modulation (jitter) and amplitude modulation (shimmer). Frequency perturbations monotonically increased (approximately 1%) with target frequency. Amplitude deviations will be discussed as they relate to the frequency modulation.

Una visión integrada del diasistema fonológico catalán

This paper describes the phonological inventor of Catalan system under the conditions established by a set of abstract structures deductively obtained. It is supposed that a natural language uses as many phonological features as archiphonemes we need postulate for it, and by doing so, each phoneme can be caracterized by only two features. No matter its methodological complexity, such a procedure solves some classical problems of Catalan phonology, especially those of neutral vowel, sibilant clusters, and glides.

Neutral vowels and the autosegmental analysis of Hungarian vowel harmony

In this paper it is shown that the behavior of neutral vowels in Hungarian, i.e. vowels that do not undergo vowel harmony, can he readily accounted for in an autosegmental analysis of vowel harmony. This analysis also allows for an insightful account of lexical exceptions to Hungarian vowel harmony (a much debated issue) without making use of absolute neutralization and ad hoc word-internal grammatical boundaries.

The effects of antagonistic gestures on temporal and amplitude parameters of anticipatory labial coarticulation.

Anticipatory lip rounding before neutral versus antagonistic vowel environments was studied in a test of the compatibility notion of a look-ahead scanning mechanism (Henke, 1967). Electromyograms were obtained from orbicularis oris superior, an agonist for lip rounding, and risorius, an agonist for lip spreading, from three normal speakers repeating the nonsense disyllables /tiku/, /taku/, /tuki/, /kikstu/, and /kakstu/ twenty times each. When the syllable-initial vowel was the lip spread vowel /i/, oris activity began earlier and with greater force than for syllable-initial /a/ contexts across all speakers. This finding argues for a revision of a key assumption underlying the look-ahead scan model for anticipatory lip rounding. The encoding program for speech exhibits temporal and amplitude adjustments in its coarticulatory behavior to accommodate in an active way contradictory neuromuscular and biomechanical conditions. These results are discussed with reference to current models of speech production.

Effects of stimulus ear presentation on the voice reaction time of adult stutterers and nonstutterers

The initial and optimum voice reaction times (VRT) to auditory stimuli presented separately to the left and right ears of ten adult stutterers and ten nonstutterers was investigated. Subjects initiated the neutral vowel sound /Λ/ in response to one hundred 4000 Hz tones of 2.5 sec in duration. The silent intervals between the tones varied randomly. The stimulus cues were divided into five equal response sets of 20 tones each with 10 tones in each set being presented to the right ear and 10 tones to the left ear alternating back and forth. No significant differences were reported between the VRTs for cues presented to the left or right ears for either group. However, the stutterers exhibited voice reaction times which were significantly longer and more variable than those of the nonstutterers. The between- group differences were observed for what appeared to be the “optimum” level of voice initiation for the experimental task. These results lend to the speculative hypothesis that the observed difficulty for adult stutterers to promptly and consistently initiate vocalization may in part be attributable to inherent rather than learned factors.

The effect of deafness on the generation of voice

The glottal source characteristics of 20 hearing‐impaired adolescents were investigated with a reflectionless tube [R. B. Monsen and A. M. Engebretson, J. Acoust. Soc. Am. 62, 981–993 (1977)]. Speakers phonated a neutral vowel in normal and soft voice and in polysyllabic words with primary stress on one syllable. These data were analyzed and compared with data from normal male and female adults and with glottal waves generated by the Flanagan‐Ishizaka two‐mass model of vocal‐fold vibration. It was found that (1) the glottal pulse produced by the deaf subjects, when considered as an isolated entity, is comparable to that of normal‐hearing speakers, although certain phonation abnormalities, such as diplophonia and creaky voice, occur quite commonly among the deaf subjects; and (2) the change of the glottal source over time, as controlled by the interaction of vocal‐fold tension and subglottal air pressure, is a primary deviant factor for many of the deaf speakers, and in particular the use of vocal‐fold tension as a frequency changing mechanism seems impaired by the subjects' deafness. [Supported by NINCDS Grant 03856.]

On the Relationship between Formant Bandwidths and Vocal-Tract Shape Features

Theoretical and experimental studies by Borg, Pinson, Schroeder, Mermelstein, and Heinz have demonstrated that the formant frequencies of vocalic speech sounds do not uniquely determine the area function of the vocal tract. For instance, a constant area function corresponding to an idealized neutral vowel can be perturbed in infinitely many ways without changing the formant frequencies, provided that the perturbing function is sufficiently “small” and is symmetric about the midpoint of the glottis-to-lip midline. It will be shown that the acoustic import of the frequency invariant perturbations is systematically reflected in the bandwidths of the formants. This claim will be supported by data collected from model experiments in which the pressure response at the lips to a velocity impulse excitation at the glottis is computed for a variety of vocal tract shapes and analyzed by means of the fast Fourier transform. The frequencies and bandwidths of the formants observed in the amplitude spectra thus computed are determined through an analysis-by-synthesis technique and related to certain features of the perturbations. The results shed some light on the possible role of the band-widths in speech perception.

Listener judgements of fluency: The effect of part-word CV repetitions and neutral vowel substitutions

Three groups of forty subjects listened individually to one of four recordings of voluntary disfluencies. Disfluencies took the form of either one or two-unit repetitions and contained either the correct vowel or the stressed neutral vowel in place of the correct vowel. Each listener rated the speaker on a five point scale where 1 represented the most fluent rating and 5 represented the least fluent rating. Results indicated no significant difference in the ratings of the four recordings. Overall severity ratings decreased as training in speech pathology increased. Graduate students in Speech Pathology rated three of the four tapes as being significantly more fluent than the non-Speech Pathology undergraduates. While Speech Pathology graduate students tended to rate the four tapes as being more fluent than did the Pre-professional Speech Pathology students, only one tape was rated significantly different.

Le phonème indéterminé

Abstract The term ‘neutral vowel’ should apply to an acoustic-articulatory description, and the term, ‘indeterminate vowel’ to a description of a phonological system. Three applications of the vowel [É™] can be observed: (1) in partial systems wherein it has no distinctive value; (2) in systems in which it is negatively determined, with a very low phonological outpout; (3) in systems in which the vowel [É™] plays the same distinctive role as the other phonemes.

On the Applications of Distinctive Features to Speech Recognition

The value of a distinctive-feature description of the speech process at the level of linguistic studies involving grammar, syntax, morphology, phonology, etc., has already been demonstrated. The value of the distinctive-feature approach to automatic-speech recognition, compared to other descriptions of the speech process, has heretofore been only partially examined. A computer-recognition scheme, based on the goal of tracking the distinctive features of speech, is presented. The program provides a complete vowel-recognition logic, which compares favorably with the behavior of a panel of listeners when both are presented with the speech of a single speaker. Methods are presented by which single vowel, diphthong, and possible triphthong recognition is produced from the speech spectral data. Details are given on the recognition of the short or lax vowels for which formant transitions and variations are often extreme. Also discussed is the possible effect of speech tempo on a proposed tense/lax boundary and its implications. Preliminary evidence suggests that Fant's neutral vowel may take on fundamental phonetic as well as articulatory significance. [Research reported in this paper sponsored by the U. S. Air Force Cambridge Research Laboratories, Office of Aerospace Research, under contract AF 19(628)305.]

COMPARING THE PROSODIC FEATURES OF ENGLISH, GERMAN, SPANISH AND FRENCH

By-Delattre, Pierre Comparing the Vocalic Features of English, German, Spanish, and French. Pub Date 64 Note-28p. Journal CitIRAL; v2 n2 p71-97 1964 EDRS Price MF -S0.25 HC-S1.50 Descriptors-Acoustic Phonetics, Articulation (Speech), Charts, *Comparative Analysis, English, French, German, Illustrations, Language Instruction, Language Research, Linguistics, *Modern Languages, Phonetic Analysis, * P hone tic s, Spanish, Tables (Data), *Vowels Phonetic characteristics of English, German, Spanish, and French vowels are compared and contrasted in this third of a series of articles on general phonetic characteristics of American. English. Much attention is given to acoustic and articulatory description. Vowel distribution, frequency, and duration are discussed. New vowel sounds for the speaker of English are identified in the other languages. Neutral vowel position, loss of vowel color, consonant anticipation, diphthongization, vowel color and syllable type, attack and release, and nasality are considered. For related documents see FL 000 782 and FL 000 784. (AF)