Distribution Of Phonemes Research Articles

Crossmodal correspondences in natural language: Distribution of phonemes and consonant-vowel patterns in Serbian words denoting round and angular objects

Since described by Köhler more than half a century ago, phonetic–iconic correspondences have been demonstrated in a series of studies showing remarkable consistency in matches of pseudowords containing specific types of phonemes (e.g., Maluma or Takete) with rounded and angular shapes. If the effect found in these experiments reveals something about processes involved in natural language interpretation, we should expect similar association between phonological properties of objects’ labels and their perceptual properties to exist in natural language as well. However, results of the studies testing this effect in natural language are rather inconsistent and sometimes even contradictory. The aim of the present study was to test whether the distribution of phonemes and consonant-vowel patterns, previously found in pseudowords participants produced for the abstract visual patterns (Jankovic and Markovic, 2000, Perception 29 ECVP Abstract Supplement), could be found in the words of natural language. For 1066 nouns denoting round and angular shapes extracted from the Corpus of Serbian Language, distribution of phonemes and consonant-vowel patterns were analyzed. Results showed that words of Serbian language denoting sharp and rounded objects show similar patterns of phoneme and consonant-vowel distributions as those found in pseudowords produced for sharp and rounded visual stimuli, and therefore provide further evidence for cross-modal correspondences in natural language. These findings were discussed in the light of the role crossmodal correspondences can have in the natural language acquisition.

‘Standard English’ as spoken in Shetland's capital

ABSTRACT: This paper deals with the local accent of Standard English spoken in Lerwick, the main town or capital in the Shetland Islands, in contrast to previous research on Shetland speech which focused on the local Scots dialect, commonly referred to as ‘Shetland dialect’. The account is based on data collected for a phonological study, in which a judgment sample of bidialectal, middle‐aged, middle‐class speakers was obtained through a network model. It is also informed by data collected in an earlier regional survey, as well as observations made over a ten year period. An analysis is presented of systemic, distributional and realizational aspects of the vowel and consonant systems. It may be concluded that the accent displays a number of localized features, especially regarding phonetic realization, but also standardized features, including various consonant contrasts and the lexical distribution of phonemes. An overview is also provided of grammatical and lexical features, which further supports the suggestion that the speech accounted for constitutes a form of (Scottish) Standard English.

Classification of intended phoneme production from chronic intracortical microelectrode recordings in speech-motor cortex.

We conducted a neurophysiological study of attempted speech production in a paralyzed human volunteer using chronic microelectrode recordings. The volunteer suffers from locked-in syndrome leaving him in a state of near-total paralysis, though he maintains good cognition and sensation. In this study, we investigated the feasibility of supervised classification techniques for prediction of intended phoneme production in the absence of any overt movements including speech. Such classification or decoding ability has the potential to greatly improve the quality-of-life of many people who are otherwise unable to speak by providing a direct communicative link to the general community. We examined the performance of three classifiers on a multi-class discrimination problem in which the items were 38 American English phonemes including monophthong and diphthong vowels and consonants. The three classifiers differed in performance, but averaged between 16 and 21% overall accuracy (chance-level is 1/38 or 2.6%). Further, the distribution of phonemes classified statistically above chance was non-uniform though 20 of 38 phonemes were classified with statistical significance for all three classifiers. These preliminary results suggest supervised classification techniques are capable of performing large scale multi-class discrimination for attempted speech production and may provide the basis for future communication prostheses.

Phonetically Balanced Text Corpus Design Using a Similarity Measure for a Stereo Super-Wideband Speech Database

In this paper, we propose a text corpus design method for a Korean stereo super-wideband speech database. Since a small-sized text corpus for speech coding is generally required for speech coding, the corpus should be designed to comply with the pronunciation behavior of natural conversation in order to ensure efficient speech quality tests. To this end, the proposed design method utilizes a similarity measure between the phoneme distribution occurring from natural conversation and that from the designed text corpus. In order to achieve this goal, we first collect and refine text data from textbooks and websites. Next, a corpus is designed from the refined text data based on the similarity measure to compare phoneme distributions. We then construct a Korean stereo super-wideband speech (K-SW) database using the designed text corpus, where the recording environment is set to meet the conditions defined by ITU-T. Finally, the subjective quality of the K-SW database is evaluated using an ITU-T super-wideband codec in order to demonstrate that the K-SW database is useful for developing and evaluating super-wideband codecs.

Development of a script of phonemically balanced monosyllable lists of Mandarin–Chinese

ObjectivesTo develop a set of monosyllable audiometry test material that can be used in speech recognition testing in Mandarin Chinese. Materials and MethodsIn an effort to improve reliability and efficiency in clinical practice, a set of 30 test lists of 25 monosyllable test items each was designed with consideration of the following: length of the list, phonemical balance (PB), word utility frequency, coverage of common words, and equivalency among lists. The 30 lists contained 750 monosyllabic words and were constructed on the basis of Chinese phoneme distribution probability summarized in the Manual of Acoustic. The occurrence incidence of the 22 consonants, 36 vowels, and 4 tones of the Chinese Mandarin characters in the 30 test lists were calculated. To achieve PB, 489 monosyllables were selected from the 2500 most commonly used Chinese characters and the 4000 most commonly used Chinese phrases to compile the 30 monosyllable test lists using a computer algorithm with manual adjustment. ResultsThirty phonemical balanced word lists of common Mandarin Chinese monosyllabic words were compiled. Each list consisted of 25 monosyllables. A total of 489 phonemes (consonants, vowels and tones) were included in the set. ConclusionsThis set of lists can potentially be used as the basic lists for future development of Mandarin PB monosyllable speech test materials.

The Analysis of Rhythm in Kim So-Wol's Poems -Concerned with distribution of phonemes-

The Analysis of Rhythm in Kim So-Wol's Poems -Concerned with distribution of phonemes-

Frequency of Occurence of Phonemes in American Spanish

A computer aided inventory of the frequency and distribution of American Spanish phonemes is reported. The corpus, drawn from five modern plays, contained 163,861 phonemes in 74,460 syllables and 43,306 words. First single elements were counted. The seven first sounds /e a o s n r i/ account for 67.5% of the total occurrences. The relative distribution of phonemes in the most frequent syllables was calculated. Higher scores correspond to /a e o/ and /s n d k t/. All phonemes were distributed according to their initial and final syllable and words position.

The secret is in the sound: from unsegmented speech to lexical categories

When learning language, young children are faced with many seemingly formidable challenges, including discovering words embedded in a continuous stream of sounds and determining what role these words play in syntactic constructions. We suggest that knowledge of phoneme distributions may play a crucial part in helping children segment words and determine their lexical category, and we propose an integrated model of how children might go from unsegmented speech to lexical categories. We corroborated this theoretical model using a two-stage computational analysis of a large corpus of English child-directed speech. First, we used transition probabilities between phonemes to find words in unsegmented speech. Second, we used distributional information about word edges--the beginning and ending phonemes of words--to predict whether the segmented words from the first stage were nouns, verbs, or something else. The results indicate that discovering lexical units and their associated syntactic category in child-directed speech is possible by attending to the statistics of single phoneme transitions and word-initial and final phonemes. Thus, we suggest that a core computational principle in language acquisition is that the same source of information is used to learn about different aspects of linguistic structure.

Quantifying Dialect Similarity by Comparison of the Lexical Distribution of Phonemes

This paper describes a new method for quantifying the similarity of the lexical distribution of phonemes in different varieties of a language (in this case English). In addition to introducing the method, it discusses phonological problems which must be addressed if any comparison of this sort is to be attempted, and applies the method to a limited data set of varieties of English. Since the method assesses their structural similarity, it will be useful for analysing the historical development of varieties of English and the relationships (either as a result of common origin or of contact) that hold between them.

Sonidos, Fonemas, Morfemas (review)

Reviewed by: Sonidos, fonemas, morfemas by Juan A. Hasler Carolina González Sonidos, fonemas, morfemas. By Juan A. Hasler. Cali, Colombia: Universidad del Valle, 2003. Pp. 249. ISBN 9586702596. Sonidos, fonemas, morfemas (Sounds, phonemes, morphemes) is a manual of universal phonetics written in Spanish for linguistics and philology students. For this reason, the linguistic discussion is supplemented by philological comments on many of the linguistic terms used. It is based on Juan A. Hasler’s class notes from the 1980s (99) and divided into two sections, the first theoretical and the second more practical: ‘Sonidos’ (5–96) and ‘Fonemas, morfemas’ (97–249). The first section, ‘Sonidos’, describes and explains the range of sounds and supra-segmental phenomena that are found in languages, with plenty of examples from diverse language families. This section also discusses to a large extent which phonetic symbols should be used. H advocates freedom of transcription provided that conventions are clearly stated by the analyst (25). This section ends with an overview of the materials covered as well as two appendices. The first appendix (72–76) elaborates on some of the issues presented in the previous pages, while the second [End Page 949] appendix (77–96), written in collaboration with Roberto Ruiz, proposes a mathematical model of phonetic features and their combination. ‘Fonemas, morfemas’ consists of a series of exercises that focus on examining linguistic inventories to find out the distribution of phonemes, allophones, allomorphs, and suprasegmental features. These exercises gradually increase in complexity and are almost entirely based on real languages (for the most part unidentified in the book); most of these exercises are explained in detail by the author. The book ends with a brief thematic index covering both sections. The book has an extensive phonetic coverage and provides a wealth of materials suitable for introductory phonology and phonetic courses or self-study. However, it could be improved on a number of fronts. There is a profusion of typographical errors, especially in the first section (the brief list of errata included in the book is helpful, but not sufficient). References, when provided, are not very current, and citations are often vague, failing to include authors and titles of studies in many cases. A revision of the text, eliminating discussion on matters such as which symbols are best for use in typing machines, would make this manual more appealing for a wider audience. Carolina González Florida State University Copyright © 2006 Linguistic Society of America

The effect of distributional information on children’s use of phonemic contrasts

Several recent experiments indicate that, when learning words, children are not as sensitive to phonemic differences (e.g., /d/ vs. /t/) as they are in discrimination tasks [Pater, J., Stager, C. L., & Werker, J. F. (2004). The perceptual acquisition of phonological contrasts. Language, 80, 384–402; Stager, C. L., & Werker, J. F. (1997). Infants listen for more phonetic detail in speech perception than in word-learning tasks. Nature, 388, 381–382]. In particular, 14-month-olds who can hear the difference between similar words such as “daw” and “taw” treat these labels as interchangeable when they refer to newly learned objects. We hypothesize that experience with the distribution of phonemes in their native language helps children use these phonemic distinctions in a word-learning task. Children receive distributional information when they experience different phonemes in different words. To test this hypothesis, we exposed children to two kinds of words: words in which phonemes occurred in very similar contexts, and words in which phonemes occurred in dissimilar contexts. Children who experienced phonemes in dissimilar contexts were more successful in using the contrast between the phonemes in a word-learning task.

Capacity and complexity of HMM duration modeling techniques

The ability of a standard hidden Markov model (HMM) or expanded state HMM (ESHMM) to accurately model duration distributions of phonemes is compared with specific duration-focused approaches such as semi-Markov models or variable transition probabilities. It is demonstrated that either a three-state ESHMM or a standard HMM with an increased number of states is capable of closely matching both Gamma distributions and duration distributions of phonemes from the TIMIT corpus, as measured by Bhattacharyya distance to the true distributions. Standard HMMs are easily implemented with off-the-shelf tools, whereas duration models require substantial algorithmic development and have higher computational costs when implemented, suggesting that a simple adjustment to HMM topologies is perhaps a more efficient solution to the problem of duration than more complex approaches.

From phonemes to meaning: John Steinbeck's work examined.

Two novels were selected from two distinct periods in John Steinbeck's career, and text samples from each were pooled according to era. Using quantitative measures of text emotion, between-sample and norm-relative comparisons were computed to evaluate a priori expectations regarding differences in the use of implicit emotional information. These hypotheses reflected abstract characteristics that transcend the focus of a single novel. Analysis of word-based measures did not yield the between-sample differences expected. Rather, the hypotheses were supported by distinct patterns of phoneme distribution. The findings confirm and extend previous work by demonstrating that authors of prose manipulate the emotional quality of phonemes, and that the associated patterns can occur at a more abstract level than previously demonstrated.

Correlation between the Perceptual Rating of Speech in Dutch Patients with Velopharyngeal Insufficiency and Composite Measures Derived from Mean Nasalance Scores

Objective: To evaluate the potential clinical use of composite measures derived from mean nasalance scores. Procedure: Speech samples with a normal distribution of phonemes (normal text, NT) and speech samples free of nasal consonants (denasal text, DT) of 43 patients with perceived hypernasality were used. The overall grade of severity, hyperrhinophonia, audible nasal emission, misarticulations associated with velopharyngeal insufficiency and intelligibility were perceptually rated on separate visual analog scales. Mean nasalance scores were computed by the Nasometer for the same speech samples on which the perceptual ratings were performed. From the mean nasalance scores computed for the NT and DT passages the difference and the quotient were calculated. The advantage could be that the derived measures provide some normalization with regard to the performance of the individual speaker. Spearman correlation coefficients were computed between these composite measures and the perceptually rated parameters. The results were compared with the correlation coefficients between the mean nasalance scores and the ratings. Setting: The Institute of Phoniatrics, Utrecht University Hospital, The Netherlands. Results: The correlations between the composite measures and the perceptual ratings were generally lower than the correlations between mean nasalance scores and the ratings. Conclusion: Normalization of the nasalance scores did not enhance the correlation with the perceptual ratings in this study.

The Effects of Local Phonetic Contrasts in Readers' Responses to a Short Story

The sound of the language in a literary text is often thought to contribute to its meaning. We hypothesize that this is due not to fixed or universal phoneme properties, as theories of phonetic symbolism have supposed, but to the use of local phonetic contrasts to elicit meaning. Writers may set an overall range of phonetic tones that are distinctive to a particular text and then introduce significant variations to achieve local effects. In the present study, an analysis of phoneme distributions developed by Miall (2001) and an approach to phonetic symbolism developed by Whissell (1999, 2000a, 2000b) were applied to a Katherine Mansfield short story. Readers' responses to the story were obtained using Semantic Differential ratings. The findings show the influence of phonetic patterns consistent with the hypothesis that phonemic contrasts elicit local changes in feeling tone. The effects of phonetic symbolism, while evident, were much less pronounced.

The Correlation Between Nasalance and a Differentiated Perceptual Rating of Speech in Dutch Patients With Velopharyngeal Insufficiency

Abstract Objective: The correlation between the nasalance score and the perceptual rating of several aspects of speech of speakers with velopharyngeal insufficiency (VPI) by six speech-language pathologists was evaluated. Procedure: The overall grade of severity, hypernasality, audible nasal emission, misarticulations, and intelligibility were rated on visual analog scales. Speech samples with a normal distribution of phonemes (normal text [NT]) and those free of nasal consonants (denasal text [DT]) of 43 patients with VPI were used. Mean nasalance scores were computed for the speech samples, and Spearman correlation coefficients were computed between the mean nasalance score and the five parameters of the differentiated rating. Setting: The Institute of Phoniatrics, Utrecht University Hospital, The Netherlands. Results: The correlation coefficient between the mean nasalance and the perceptual rating of hypernasality ranged among judges from .31 to .56 for NT speech samples and .36 to .60 for DT speech sa...

The correlation between nasalance and a differentiated perceptual rating of speech in Dutch patients with velopharyngeal insufficiency.

The correlation between the nasalance score and the perceptual rating of several aspects of speech of speakers with velopharyngeal insufficiency (VPI) by six speech-language pathologists was evaluated. The overall grade of severity, hypernasality, audible nasal emission, misarticulations, and intelligibility were rated on visual analog scales. Speech samples with a normal distribution of phonemes (normal text [NT]) and those free of nasal consonants (denasal text [DT]) of 43 patients with VPI were used. Mean nasalance scores were computed for the speech samples, and Spearman correlation coefficients were computed between the mean nasalance score and the five parameters of the differentiated rating. The Institute of Phoniatrics, Utrecht University Hospital, The Netherlands. The correlation coefficient between the mean nasalance and the perceptual rating of hypernasality ranged among judges from .31 to .56 for NT speech samples and .36 to .60 for DT speech samples. Only small differences were found between speech pathologists with and without expertise in cleft palate speech. The rating of the overall grade of severity appeared to correlate quite well with the rating of the intelligibility (r(NT) = .77, r(DT) = .79). Lower correlation coefficients, ranging from .34 to .71, were found between overall grade of severity and hypernasality, audible nasal emission, and misarticulations. A low correlation between the nasalance and the perceptual rating of hypernasality was found. The parameter overall grade of severity appeared to be determined mainly by the parameter intelligibility. Expertise in rating of cleft palate speech does not guarantee a high correlation between instrumental measurement and perceptual rating.

Phoneme-group specific octave-band weights in predicting speech intelligibility

In an earlier study we derived robust frequency-weighting functions for prediction of the intelligibility of short nonsense words. These frequency-weighting functions are applied for prediction of intelligibility such as with the speech transmission index (STI). Six independent experiments revealed essentially similar frequency-weighting functions for the prediction of the nonsense word scores with respect to signal-to-noise ratio and gender [Speech Communication 28 (1999) 109]. Although the frequency weightings do not vary significantly for signal-to-noise ratio or gender, other studies have shown that using different types of speech material (i.e., nonsense words, phonetically balanced words and connected discourse) resulted in quite different frequency-weighting functions. This may be related to the distribution of specific phonemes in the test material. In order to obtain a more generic description of the frequency weighting, four relevant groups of phonemes were identified. In situations with reduced intelligibility, a small confusion rate of the phonemes between the groups and a high confusion rate of the phonemes within each group was observed. For each group a specific frequency-weighting function and a good prediction of the phoneme group scores could be obtained. It was shown that from these (weighted) phoneme group scores, word scores could be predicted with a prediction accuracy of ca. 4% (this corresponds to a signal-to-noise ratio of about 1 dB). Hence, this method provides a more generic way to predict intelligibility scores for different types of speech material.

Sound and Emotion in Given Names

AbstractAn analysis of the distribution of phonemes in men's and women's names confirmed several past findings, e.g., women's names were more variable and longer, contained more vowels, and were more likely to end with a vowel, especially schwa. Assigning emotional character to individual phonemes revealed that emotional information was encoded in name sounds. Men's names included significantly more cheerful, active, nasty, and unpleasant phonemes while women's names contained more soft, pleasant, passive, and sad phonemes. Phonemic information successfully classified sex. Finally, common androgynous names were found to be emotionally feminine in their distribution of phonemes.

Phonosymbolism and the emotional nature of sounds: evidence of the preferential use of particular phonemes in texts of differing emotional tone.

This article addresses the emotional meaning (phonosymbolism) of the most basic unit of language--the phoneme. Language excerpts from many sources were transcribed phonetically with the help of a computer program. The distributions of phonemes in different sources (song lyrics, poetry, word lists, advertisements) were correlated with the emotionality of the language along two dimensions (activation, evaluation) which had been rated by the Dictionary of Affect using another computer program. Significant results characterized all phases of the analysis. Phonemes were distributed differently in different language samples. Frequency of phonemes in a language sample as well as frequency of phonemes in individual words were correlated with emotion. For example, the phoneme /l/ (the one appearing twice in the word lullaby) was used more often in pleasant language samples, in soft or tender language samples, and in passive words. The phoneme /r/ (appearing twice in the word roar) was found more often in unpleasant words and in active words. Possible sources of the relationship between sound production and emotion are discussed.