Similarity Of Consonants Research Articles

The similar rates of occurrence of consonants across the world's languages: A quantitative analysis of phonetically transcribed word lists

Via an analysis of nearly 7000 phonetically transcribed word lists representing every major linguistic family, this study examines the frequency with which languages use particular consonants. The results suggest that there is an underlying global similarity in the frequency of occurrence of these consonants. Some consonants are uncommon across the word lists, and do not occur frequently even in those languages in which they are used. In contrast, a few other consonants represent a large portion of sounds in the word lists, across regions and linguistic families. The analysis quantifies the pervasiveness of such sounds that are known to be, impressionistically, common in speech. A new method captures the overall similarity of consonants’ rates of occurrence across languages and families. This method offers one means of evaluating the extent to which individual languages or families deviate from, or adhere to, the typical patterns of consonant usage. It is suggested that the crosslinguistic similarity of consonant usage, in terms of rates of occurrence in word lists, likely owes itself to previously documented factors like the relative ease of articulation of some sounds. This new evidence suggests that the role of ease of articulation in shaping speech may be more influential than generally assumed, though other explanatory factors are also likely at work and these patterns require further exploration with more robust intralinguistic phonetic corpora.

Similarity Index for Sound-alikeness of Drug Names with Pitch Accents

Drug name similarity is one of major reasons of medical accidents. In order to prevent from the accidents, one of the best ways is to avoid approving drugs that has the names similar to that of existing drugs. It is well-known that there are two kinds of drug name similarity, look-alikeness and sound-alikeness. Nabeta et. al. proposed a look-alikeness similarity index,which excludes the sound-alikeness. Though, in Japan, oral prescription is basically prohibited, emergent situation can force a doctor to prescribe orally. In such a situation, medical accidents can occur.In this study, we proposed a sound-alikeness similarity index based on quantitative similarity of consonants. The consonant similarity was proposed based on The International Phonetic Alphabet (IPA). Overall drug name similarity is calculated based on Letter Sequence Kernel (LSK). The similarity calculation method takes account of the effect of plural pitch accents. We divided a drug name into some pieces at the position where a pitch accent changes, applied LSK to each of them, and combined them to obtain the value of the similarity index. The similarity index proposed in this study achieved relatively high correlation to the results of our experiment, r ≃ 0.8.

Consonant confusability and its relation to phonological similarity

Gradient similarity avoidance patterns in Bengali echo reduplication suggest that the most similar consonants to /t/ are, in order, /t, th, d, t, s, th, k, .../. Converted to confusability, this ranking predicts that aspiration is most confusable, followed by voicing, minor place, continuancy, major place, and sonority. To confirm this, 24 native speakers identified syllables masked with babble, noise, or nothing (“clear”). Results indicate that confusability reflects similarity as predicted by avoidance patterns. In clear speech, most errors involved voicing and aspiration. Other errors reflected dialect-specific alternations. Noise introduced the percept of a loud burst: fricatives were often heard as affricates, non-alveolar coronals as alveolars, and non-coronals as coronals. Babble largely resembled noise. This pattern suggests that voicing is the most confusable feature, followed by aspiration, minor place and continuancy, major place, and lastly sonority. This ranking seems Bengali-specific, as studies of English find place to be significantly more confusable than manner and voicing. These results suggest that at least for Bengali, phonological alternations and perceptual confusability are argued to be better representations of how speakers judge similarity, rather than patterns in the lexicon or metrics such as shared natural classes metric of Frisch et al. (2004).

Consonant confusability and its relation to phonological dissimilarity

Consonant similarity can be measured indirectly through a language’s phoneme inventory, lexicon (e.g. cooccurrence restrictions), or phonology (e.g., processes that take similarity or dissimilarity into account). It can also be measured more directly as confusability in a perception task. Thus far, consonant similarity in Bengali has only been measured indirectly, through the inventory, lexicon, and phonology. Previous studies [Khan (2006)] claim that Bengali speakers judge the similarity of consonants in echo reduplication, where the initial consonant of the base is systematically replaced with a phonologically dissimilar consonant in the reduplicant, e.g., kashi “cough” > kashi-tashi “cough, etc.” but thonga “bag” > *thonga-tonga > thonga-fonga “bags, etc.”). This measurement of similarity assumes a set of features assigned language-specific weights; for example, [voice] is weighted more heavily that [spread glottis], to explain why speakers treat the pair [t, th] as more similar than the pair [t, d]. But does the measurement of similarity inherent in the echo reduplicative construction correspond directly to the relative perceptibility of different consonant contrasts? The current study examines data collected in a perception experiment, comparing the relative confusability of Bengali consonants produced in noise with the claims of phonological notions of similarity associated with echo reduplication.

Statistical learning in reading: Variability in irrelevant letters helps children learn phonics skills.

Early reading abilities are widely considered to derive in part from statistical learning of regularities between letters and sounds. Although there is substantial evidence from laboratory work to support this, how it occurs in the classroom setting has not been extensively explored; there are few investigations of how statistics among letters and sounds influence how children actually learn to read or what principles of statistical learning may improve learning. We examined 2 conflicting principles that may apply to learning grapheme-phoneme-correspondence (GPC) regularities for vowels: (a) variability in irrelevant units may help children derive invariant relationships and (b) similarity between words may force children to use a deeper analysis of lexical structure. We trained 224 first-grade students on a small set of GPC regularities for vowels, embedded in words with either high or low consonant similarity, and tested their generalization to novel tasks and words. Variability offered a consistent benefit over similarity for trained and new words in both trained and new tasks.

Confusability of Bengali consonants and its relation to phonological dissimilarity

Language-specific consonant similarity can be measured indirectly by looking at the phoneme inventory, the lexicon (e.g. cooccurrence restrictions), or the phonology (e.g. processes that take the notion of similarity of dissimilarity into account). A more direct approach involves the use of the confusion matrix. For Bengali, thus far, consonant similarity has only been measured indirectly, through the lexicon and phonology. Previous studies (Khan 2006, 2012) claim that Bengali speakers judge the similarity of consonants in echo reduplication (similar to English doctor-schmoctor), where the initial consonant of the base is systematically replaced with a phonologically dissimilar consonant in the reduplicant. This measurement of similarity assumes a set of features assigned language-specific weights; for example, [voice] is weighted more heavily that [spread glottis], to explain why speakers treat the pair [t, th] as more similar than the pair [t, d]. But does the measurement of similarity inherent in the echo reduplicative construction correspond to the relative perceptibility of different consonant contrasts? The current study compares the relative confusability of Bengali consonants produced in noise with the claims of phonological notions of similarity associated with echo reduplication.

Nasal and oral consonant similarity in speech errors: Exploring parallels with nasal consonant harmony

Previous research has found that ‘similar’ sounds interact in phonological nasal consonant harmony, wherein certain consonants become nasals when the word contains a nasal (e.g., Kikongo: /-kun-idi/ → [-kun-ini] ‘planted’). Across languages, stops and approximants are chiefly affected, especially voiced consonants and ones that match in place of articulation with the nasal. Three experiments investigated whether a parallel occurs in consonants showing greater likelihood to interact in speech errors with nasals. The experiments, which elicited errors using the SLIPS technique with English speakers, revealed the following asymmetries in consonants’ participation in errors with nasals: (i) voiced stops (b, d) > voiceless stops (p, t), (ii) voiced stops with same place > voiced stops with different place, and (iii) approximants (r, l) > voiceless stops (p, t). These correlate with preferentially affected segments in nasal consonant harmony. The data support a uniform phonological similarity scaling for nasal-oral consonants across phonological harmony processes and speech errors. Further, they are consistent with theoretical proposals that consonant harmony has functional origins in facilitating language production.

Phonetic structure is similar across auditory and vibrotactile speech perception

Vibrotactile vocoders have long been viewed as a possible method for conveying speech information to deaf perceivers. However, modest levels of speech transmission have led to numerous alternative notions concerning limitations of vibrotactile devices. This study investigated whether perceived acoustic phonetic structure is preserved in the transformation of the F2 speech region via a linear vibrotactile vocoder array. In experiment 1, naive hearing subjects discriminated pairs of vibrotactile word stimuli immediately before and after 2 h of practice. At both times, relative vibrotactile discrimination levels were well predicted by identification of phonemes presented by an analogous F2 acoustic vocoder, suggesting that structure is preserved. In experiment 2, subjects separately rated auditory and vibrotactile consonant similarity of stimuli presented by auditory and vibrotactile vocoders, respectively. Analysis of phonological feature information showed similarities and differences in perceptual structure across modalities. Vibrotactile stimuli were rated to be more similar to each other than were auditory stimuli, suggesting that phonetic information was degraded relative to auditory information. Based on these results, current investigations are directed at methods for inducing perceptual learning of vibrotactile phonetic information. [Work supported by NIH/NIDCD 00695.]

Perceptual learning of synthetic speech

Over the course of a lifetime, listeners are likely to encounter many unfamiliar talkers speaking the listeners’ native language. Despite the fact that talkers may differ radically in the way they sound, listeners normally learn to understand the novel speech patterns of unfamiliar talkers without difficulty. Part of learning to understand an unfamiliar talker speaking a familiar language is learning to relate novel acoustic patterns in the unfamiliar speech to existing mental representations of familiar phonetic categories. This process has frequently been described in terms of modifying an existing psychological ‘‘space’’ of mental representations by shifting the focus or weight of attention to linguistically useful aspects of the speech signal. In the experiment reported here, native English listeners were trained to better understand an unfamiliar talker—a computer speech synthesizer. Training to recognize and transcribe English words resulted in a significant increase in listeners’ abilities to understand words and to identify consonants. Training also strongly influenced the distribution of attention to particular features of the speech signal, as indicated by changes in listeners’ judgments of the similarity of consonants after training. Implications for theories of perceptual learning and phonetic categorization will be discussed.

日本語子音音素の類似性 ＩＮＤＳＣＡＬと林の数量化理論第Ｉ類による分析

The purpose of this study was to identify the dimensions relevant to perception of 13 Japanese consonant phonemes. Nine subjects read aloud a list of pairs of monosyllables (C+/a/) and judged the auditory similarity of consonants to each other. Similarity responses were analyzed by INDSCAL and Hayashi's Quantification Theory I. Four distinctive features in Chomsky-Halle feature system were found to be important through the latter analysis. Three-dimensional configuration derived from the former one was interpreted on the basis of these articulatory features. Dimension 1 was interpreted as the effects upon the perception produced by both the feature [strident] and [coronal], Dimension 2 by [[-sonorant]∧[+voiced]], and Dimension 3 by both [voiced] and [sonorant]. This interpretation proved to be highly valid.

Consonant similarity judgments by normal and hearing-impaired listeners.

This investigation attempted to identify listener strategies or perceptual modes that might be adopted by hearing-impaired listeners when making similarity judgments among pairs of speech sounds. Further, an attempt was made to describe the relationship between similarity judgments and auditory confusions for such listeners. Subjects provided similarity ratings and recognition responses to consonant pairs. The resulting similarity judgments were organized into a variety of similarity matrices and analyzed via multidimensional scaling and hierarchical clustering, as well as by traditional descriptive and interpretative statistics. The analyses of the similarity ratings between consonants showed that hearing-impaired listeners apply phonemic labels to the stimuli and base their ratings on these labels rather than on the unlabeled acoustic characteristics of the speech sounds. Analysis of the recognition data indicated that those consonants which are most confused are not necessarily the most conceptually similar to the listener.

Ratings of the auditory and visual similarity of consonants: Implications for research

A set of rating data was collected which compared the visual similarity of consonants to the auditory similarity of the same letters. Analysis of the rating patterns raises important questions about the use of similarity data in drawing conclusions about mode of memory coding in a variety of situations.

Distinctive feature analysis of choice reaction times of consonantal similarity judged by children of different age groups

Choice reaction time (CRT) was used as criterion to determine strategies in judging similarity of consonants. Subject comprised thirty children of 5, 6, and 7 years with five males and females in each age group. The ensemble of English phonemes were /k, f, v, θ, ∂, s, z, ∫, e,j, w, r/ each followed by /a/. The overall CRT for 5- and 6-year olds were not different; however, they were different for the 7-year olds. The CRT for females were different from the males. A distinctive feature analysis revealed that a greater CRT was found for minimally distinct pair than any other pair. No difference was found among the CRT's associated with pairs different by two-five features. It was concluded that these children utilize distinctive features in making perceptral judgments by availing themselves with the knowledge of the redundancy aspect of the differences. The INDSCAL analysis revealed that the utilization of features and their weightings by the three age groups showed no systematic differences. The perceptual features obtained from these children's judgments had both similarities and differences of features manifested in earlier studies on adults.

Dimensions of consonant perception in normal and hearing-impaired listeners.

Judgments of consonant similarity were obtained from subjects who had normal hearing, high-frequency sensorineural hearing loss, or relatively flat sensorineural hearing loss. The individual differences model through program INDSCAL was used to derive a set of perceptual features empirically from the similarity judgments, and to group the subjects on the basis of strength of feature usage. The analysis revealed that sonorance was the dominant dimension in the similarity judgments of the subjects with high-frequency hearing losses, while sibilance tended to dominate the judgments of the subjects with flat audiometric configurations. The normal-hearing subjects tended to weight these two dimensions approximately equally. These differences in similarity judgments were observed based upon audiometric configuration, despite the fact that the two hearing-impaired groups were not unique in word-recognition ability.