Vowel Normalization Research Articles

DÎVÂNU LUGÂTİ’T-TÜRK İLE KARŞILAŞTIRMALI ALTAY TÜRKÇESİNDE İKİNCİL UZUN ÜNLÜLERİN VARLIĞI

There are 8 normal /a/, /e/, /ı/, /i/, /o/, /ö/, /ö/, /u/, /ü/ and 8 long vowels /aa/, /ee/, /ıı/, /ii/, /oo/, /öö/, /uu/, /üü/ in Altai Turkish. The frequency of use of /aa/, /ee/, /oo/, /öö/, /uu/, /üü/ is wider, while the frequency of use of /ıı/ and /ii/ is narrower, but there are examples where they are used. Altai Turkish has a significant number of long vowel formations that can be analysed in the historical process. Long vowel is expressed by the fact that the articulation time is longer than the normal vowel. Long vowels, which are handled under two headings according to their formation times, are divided into primary (‘original’) and secondary long vowels. Primary long vowels in Altai Turkish did not continue to exist regularly in the historical process like Yakut Turkish and Turkmen Turkish, but in this study, firstly, the examples of primary long vowels that have been preserved from the past to the present, identified as a result of the scanning study from the Altai Turkish dictionary, have been examined. Secondly, it is aimed to determine the existence of secondary long vowels that have formed as a result of contraction in the word structure in Altai Turkish from the Middle Turkish period to the present day, with the vocabulary found in Dîvânu Lugâti't-Türk. The aim of the study is to determine the secondary long vowels formed in the word as a result of contraction in Altai Turkish. The words identified from Altai Turkish Dictionary and Dîvânu Lugâti't-Türk vocabulary has been detected by using diachronic historical study method and scanning method. Keywords: Altai Turkish, Dîvânu Lugâti't-Türk, long vowels, primary long vowels, secondary long vowels.

A Cross-Language Comparative Study of English and German Vowel Space Area in Swiss German L1 Speakers

It is conventional to examine vowel systems by their acoustic correlates, specifically formants. In the International Phonetic Alphabet (IPA) vowel chart, vowels are placed according to their relative positions determined by the first two formant values. This two-dimensional plane makes possible the calculation of the articulatory working space of vowels. Variations in the size and shape of the vowel space exist across gender, dialects, and languages. The present study focuses on comparing the vowel space area (VSA) in English and German produced by native Swiss German speakers, attempting to shed light on whether VSAs differ significantly with different calculation methods cross-linguistically. Given the fact that absolute formant values vary tremendously across gender, Hertz values are converted to z-scores, allowing for more direct and meaningful comparisons. While vowel normalization mitigates gender-related differences within specific languages to a large extent, there remains a significant cross-linguistic gender difference. The results reveal that for 3- and 4-vowel space there is no significant effect in language but an effect is observed in gender and vowel combination on VSA, specifically in terms of individual area and overlapping ratio. This suggests that Swiss German speakers tend to utilize the same amount of vowel space in both languages, with the only difference being the expansion of English VSA in higher F2 dimension and lower F1 dimension, as observed across all calculation methods.

Evaluating normalization accounts against the dense vowel space of Central Swedish.

Talkers vary in the phonetic realization of their vowels. One influential hypothesis holds that listeners overcome this inter-talker variability through pre-linguistic auditory mechanisms that normalize the acoustic or phonetic cues that form the input to speech recognition. Dozens of competing normalization accounts exist-including both accounts specific to vowel perception and general purpose accounts that can be applied to any type of cue. We add to the cross-linguistic literature on this matter by comparing normalization accounts against a new phonetically annotated vowel database of Swedish, a language with a particularly dense vowel inventory of 21 vowels differing in quality and quantity. We evaluate normalization accounts on how they differ in predicted consequences for perception. The results indicate that the best performing accounts either center or standardize formants by talker. The study also suggests that general purpose accounts perform as well as vowel-specific accounts, and that vowel normalization operates in both temporal and spectral domains.

Normalization of nonlinearly time-dynamic vowels.

This study compares 16 vowel-normalization methods for purposes of sociophonetic research. Most of the previous work in this domain has focused on the performance of normalization methods on steady-state vowels. By contrast, this study explicitly considers dynamic formant trajectories, using generalized additive models to model these nonlinearly. Normalization methods were compared using a hand-corrected dataset from the Flemish-Dutch Teacher Corpus, which contains 160 speakers from 8 geographical regions, who spoke regionally accented versions of Netherlandic/Flemish Standard Dutch. Normalization performance was assessed by comparing the methods' abilities to remove anatomical variation, retain vowel distinctions, and explain variation in the normalized F0-F3. In addition, it was established whether normalization competes with by-speaker random effects or supplements it, by comparing how much between-speaker variance remained to be apportioned to random effects after normalization. The results partly reproduce the good performance of Lobanov, Gerstman, and Nearey 1 found earlier and generally favor log-mean and centroid methods. However, newer methods achieve higher effect sizes (i.e., explain more variance) at only marginally worse performances. Random effects were found to be equally useful before and after normalization, showing that they complement it. The findings are interpreted in light of the way that the different methods handle formant dynamics.

SPECTRAL AND TEMPORAL PROCESSING IN BALINESE VOWEL OF NON-FLUENT APHASIA

Acoustic investigation on Balinese vowel of non-fluent aphasia (NFA) has not been previously paid attention yet. Thus, this study examined the acoustic abnormalities of vowel articulation for patients with NFA. Therefore, spectral and temporal charateristic of their vowel sound are essentially searched by comparing the formant value and prosodic features with the normal vowel articulation. Speech output of two patients were observed and analyzed by using Praat and the data were described by implementing the theory of clinical phonetics and acoustics. The spectral analysis showed that the inaccurate constriction of the tongue in vowel articulation affected the range of oral tract (F2), pharynx space (F1), and shape of the lips (F3). Furthermore, the lesion in Broca’s area affected the temporal features of the sounds, such as the lower pitch, lower intensity and longer timing especially voicing environment. The vowels preceded by voiced consonants were significantly longer than those preceded by voiceless ones. So, the highly complex vowels tend to be articulated inaccurately due to articulatory implementation deficit. This finding is consistent with previous results that the spectral and temporal distortion are primarily phonetics rather than phonological planning program.

Interesting Pausal Forms in the Speech of Muslims and Christians in Kufᵘr-Kanna

The aim of this paper is to describe the system of vocalic variants in pause pertaining to speakers of Arabic in Kufᵘr-Kanna (AKK) and in this regard to determine the features that characterize the AKK. As in Nazareth, the incidence of pauses in AKK varies and depends on the content, the listener and the speaker’s intentions. In AKK I detected pausal forms in the speech of middle-aged and elderly Muslims and elderly Christians. In addition to changes in consonants and vowel quality in their speech, in pausal position final syllables also undergo other modifications as compared to the contextual forms.
 Unlike in Nazareth, four further types were identified in AKK: (1) lengthening of short vowels in final position: ‑Cv > ‑Cv̄#, -CvC > -Cv̄C#; lengthening of normal and anaptyctic short vowels in final closed syllables: -CvC#; (2) devoicing of voiced consonants in word-final position; (3) glottalization after con­sonants and vowels in word-final position; and (4) aspiration: addition of (h) in pausal position where the word ends in long vowels.
 Key words: Arabic dialects – Pausal forms – Syllables – Long vowels – Short vowels – Christians and Muslims.

The time course of normalizing speech variability in vowels

To achieve perceptual constancy, listeners utilize contextual cues to normalize speech variabilities in speakers. The present study tested the time course of this cognitive process with an event-related potential (ERP) experiment. The first neurophysiological evidence of speech normalization is observed in P2 (130–250 ms), which is functionally related to phonetic and phonological processes. Furthermore, the normalization process was found to ease lexical retrieval, as indexed by smaller N400 (350–470 ms) after larger P2. A cross-language vowel perception task was carried out to further specify whether normalization was processed in the phonetic and/or phonological stage(s). It was found that both phonetic and phonological cues in the speech context contributed to vowel normalization. The results suggest that vowel normalization in the speech context can be observed in the P2 time window and largely overlaps with phonetic and phonological processes.

Integral perception, but separate processing: The perceptual normalization of lexical tones and vowels

In tonal languages, speech variability arises in both lexical tone (i.e., suprasegmentally) and vowel quality (segmentally). Listeners can use surrounding speech context to overcome variability in both speech cues, a process known as extrinsic normalization. Although vowels are the main carriers of tones, it is still unknown whether the combined percept (lexical tone and vowel quality) is normalized integrally or in partly separate processes. Here we used electroencephalography (EEG) to investigate the time course of lexical tone normalization and vowel normalization to answer this question. Cantonese adults listened to synthesized three-syllable stimuli in which the identity of a target syllable — ambiguous between high vs. mid-tone (Tone condition) or between /o/ vs. /u/ (Vowel condition) — was dependent on either the tone range (Tone condition) or the formant range (Vowel condition) of the first two syllables. It was observed that the ambiguous tone was more often interpreted as a high-level tone when the context had a relatively low pitch than when it had a high pitch (Tone condition). Similarly, the ambiguous vowel was more often interpreted as /o/ when the context had a relatively low formant range than when it had a relatively high formant range (Vowel condition). These findings show the typical pattern of extrinsic tone and vowel normalization. Importantly, the EEG results of participants showing the contrastive normalization effect demonstrated that the effects of vowel normalization could already be observed within the N2 time window (190–350 ms), while the first reliable effect of lexical tone normalization on cortical processing was observable only from the P3 time window (220–500 ms) onwards. The ERP patterns demonstrate that the contrastive perceptual normalization of lexical tones and that of vowels occur at least in partially separate time windows. This suggests that the extrinsic normalization can operate at the level of phonemes and tonemes separately instead of operating on the whole syllable at once.

Perceptual validation of vowel normalization methods for variationist research

AbstractThe evaluation of normalization methods sometimes focuses on the maximization of vowel-space similarity. This focus can lead to the adoption of methods that erase legitimate phonetic variation from our data, that is, overnormalization. First, a production corpus is presented that highlights three types of variation in formant patterns: uniform scaling, nonuniform scaling, and centralization. Then the results of two perceptual experiments are presented, both suggesting that listeners tend to ignore variation according to uniform scaling, while associating nonuniform scaling and centralization with phonetic differences. Overall, results suggest that normalization methods that remove variation not according to uniform scaling can remove legitimate phonetic variation from vowel formant data. As a result, although these methods can provide more similar vowel spaces, they do so by erasing phonetic variation from vowel data that may be socially and linguistically meaningful, including a potential male-female difference in the low vowels in our corpus.

The ΔF method of vocal tract length normalization for vowels

Given the acoustic consequences of physiological differences between talkers, there is a practical need for effective and theoretically motivated procedures of vowel normalization to facilitate comparison of speech produced by people who differ by dialect or language. In addition, there is a question whether listeners might utilize a normalization procedure during speech perception. This paper reports the results of two studies that explore these questions—with particular focus on vocal tract length normalization. Drawing on research in speech engineering, where accurate estimates of vocal tract length are needed in some approaches to automatic speech recognition and speaker verification, a new model of vowel normalization is introduced. The model uses a direct measure of average formant spacing (the ΔF) which can be used to measure vocal tract length. The acoustic consequences of vocal tract length differences are removed from vowel measurements by scaling vowel formant measurements by ΔF. Study 1 found that this method is comparable to Nearey’s (1978) uniform normalization method, while providing an explicit vocal tract length interpretation, and a rationalized unit of measure. Study 2 found that uniform normalization measures (which let each formant serve as a noisy estimator of ΔF) improve vowel classification even with only a couple of randomly selected vowel tokens. This suggests that vocal tract length normalization could be involved in speech perception.

Vowel normalization as perceptual constancy

Vowel normalization as perceptual constancy

Vowel normalization as perceptual constancy

Vowel normalization as perceptual constancy

A Pathological Multi-Vowels Recognition Algorithm Based on LSP Feature

At present, pathological voice recognition is mainly based on the classification of pathological voice. However, almost all the researches are based on the single vowel \a samples, but few on multi-vowels. In addition, the current researches on multi-vowels recognition are mainly for normal voices, which are unsuitable for the speech recognition of normal and pathological multi-vowels simultaneously. This paper concentrates on developing an accurate and robust feature called enhanced-bark line spectrum pair (E-BLSP) to detect and classify normal and pathological multi-vowels. We explore the impact of E-BLSP feature on recognition performance and propose an effective method based on the combination of three features including E-BLSP for pathological and normal multi-vowels. In this paper, first LSP and difference of adjacent LSP (DAL) features of a vowel are extracted. Then LSP feature is warped at bark domain to get bark line spectrum pair (BLSP). In addition, then E-BLSP feature is calculated by adjusting BLSP using DAL feature. Finally, the adjusted E-BLSP feature and other two traditional features, including linear prediction cepstrum coefficient (LPCC) and mel-frequency cepstrum coefficients (MFCC) are applied to support vector machine (SVM) and deep neural network (DNN) classifiers to explore the classification performance of single feature and feature combinations for pathological and normal vowels /a/, /i/ and /u/. The results show that the highest achieved accuracies for DNN and SVM network are 98.6190% and 96.2693%, while the largest achieved area under curves (AUC) are 0.9925 and 0.9868, correspondingly with the combination of three features including LPCC, MFCC, and E-BLSP.

A Design of Low Power Electrolarynx with Glottal Modified Source

The larynx is a flexible part in the respiratory tract, a source of vibration during verbal communications. An electrolarynx or artificial larynx is a common device adopted by the laryngectomee to restore the speech. The objective of this paper is to design a low power electrolarynx using modified glottal source and perform an objective and subjective analysis of the vowel quality. In the first part, an electrolarynx circuit is designed which can drive different types of input signals. In the second part, a driving source is designed by pre-filtering the effect of neck surface. The power consumption of the prototype is measured and compared with the conventional type of electrolarynx. The vowel qualities are compared with that of normal vowels produced by a volunteer. We have found out that the glottal modified wave can reduce the power supply requirement of the electrolarynx. Loudness, quality factor and position of the formants are used as measures of quality for vowels produced by the present approach and found to be better (or comparable) than the traditional electrolarynx. When compared to existing method, the electrolarynx with glottal modified wave is power efficient and has a potential in forming a wearable device. It also has better vowel quality than the one with the conventional driving signal.

A Design of Low Power Electrolarynx with Glottal Modified Source

The larynx is a flexible part in the respiratory tract, a source of vibration during verbal communications. An electrolarynx or artificial larynx is a common device adopted by the laryngectomee to restore the speech. The objective of this paper is to design a low power electrolarynx using modified glottal source and perform an objective and subjective analysis of the vowel quality. In the first part, an electrolarynx circuit is designed which can drive different types of input signals. In the second part, a driving source is designed by pre-filtering the effect of neck surface. The power consumption of the prototype is measured and compared with the conventional type of electrolarynx. The vowel qualities are compared with that of normal vowels produced by a volunteer. We have found out that the glottal modified wave can reduce the power supply requirement of the electrolarynx. Loudness, quality factor and position of the formants are used as measures of quality for vowels produced by the present approach and found to be better (or comparable) than the traditional electrolarynx. When compared to existing method, the electrolarynx with glottal modified wave is power efficient and has a potential in forming a wearable device. It also has better vowel quality than the one with the conventional driving signal.

A regression approach to vowel normalization for missing and unbalanced data.

Researchers investigating the vowel systems of languages or dialects frequently employ normalization methods to minimize between-speaker variability in formant patterns while preserving between-phoneme separation and (socio-)dialectal variation. Here two methods are considered: log-mean and Lobanov normalization. Although both of these methods express formants in a speaker-dependent space, the methods differ in their complexity and in their implied models of human vowel-perception. Typical implementations of these methods rely on balanced data across speakers so that researchers may have to reduce the data available in the analyses in missing-data situations. Here, an alternative method is proposed for the normalization of vowels using the log-mean method in a linear-regression framework. The performance of the traditional approaches to log-mean and Lobanov normalization against the regression approach to the log-mean method using naturalistic, simulated vowel-data was investigated. The results indicate that the Lobanov method likely removes legitimate linguistic variation from vowel data and often provides very noisy estimates of the actual vowel quality associated with individual tokens. The authors further argue that the Lobanov method is too complex to represent a plausible model of human vowel perception, and so is unlikely to provide results that reflect the true perceptual organization of linguistic data.

Vocal Tract Length Normalization

The resonant frequencies of the vocal tract during vowel production convey information about the linguistic vowel intended by the talker - whether they mean to say ‘hey’ or ‘hoe’, for example - while also conveying information about the talker. One particularly salient bit of talker information that partially determines the frequencies of the vowel formants is the length of the talker’s vocal tract. Vowel formant normalization aims to remove the effects of talker differences without also removing important linguistic information. This paper presents a study of vocal tract length normalization using a new ΔF method, and compares this method to other vowel normalization methods. A key point of comparison in this study is the number of vowel tokens that are needed in order to derive a stable estimate of vocal tract length. Several of the vowel normalization methods that are most commonly used in phonetic studies are shown to need a full set of vowels in order to be reliable, while methods that derive vocal tract length information from the full acoustic spectrum are much more stable and may even provide a length-normalized representation that could be cognitively computed and used in human speech perception.

A regression approach to vowel normalization for missing and unbalanced data

Researchers investigating the vocalic systems of languages or dialects frequently employ normalization methods to minimize between-speaker variability in formant patterns while largely preserving dialectal, between-phoneme variability. One popular method, log-mean normalization, relies on estimating the logarithm of the geometric mean formant-frequency (Gs) produced by a speaker across their vocalic inventory, and then expressing the formant patterns produced by a speaker as deviations from this mean. However, in the face of missing or unbalanced data, the traditional approach to calculating Gs for a speaker will usually lead to biased estimates, which will produce artificial asymmetries in the normalized vowel spaces of different speakers. An alternative method is proposed for the estimation of Gs based on a linear-regression framework, which avoids the biases associated with traditional estimation of Gs when data is unbalanced. The regression method to normalization is described, and simulations are carried out to compare the accuracy of Gs estimates via regression to other estimation methods. Results indicate that the proposed method is substantially more accurate than the traditional approach to estimating Gs in the face of missing or unbalanced data.

Introduction to the Special Issue on Advancing Methods for Analyzing Dialect Variation

Documenting and analyzing dialect variation is traditionally the domain of dialectology and sociolinguistics. However, modern approaches to acoustic analysis of dialect variation have their roots in Peterson and Barney's [(1952). J. Acoust. Soc. Am. 24, 175-184] foundational work on the acoustic analysis of vowels that was published in the Journal of the Acoustical Society of America (JASA) over 6 decades ago. Although Peterson and Barney (1952) were not primarily concerned with dialect variation, their methods laid the groundwork for the acoustic methods that are still used by scholars today to analyze vowel variation within and across languages. In more recent decades, a number of methodological advances in the study of vowel variation have been published in JASA, including work on acoustic vowel overlap and vowel normalization. The goal of this special issue was to honor that tradition by bringing together a set of papers describing the application of emerging acoustic, articulatory, and computational methods to the analysis of dialect variation in vowels and beyond.

Neuromagnetic correlates of voice pitch, vowel type, and speaker size in auditory cortex

Vowel recognition is largely immune to differences in speaker size despite the waveform differences associated with variation in speaker size. This has led to the suggestion that voice pitch and mean formant frequency (MFF) are extracted early in the hierarchy of hearing/speech processing and used to normalize the internal representation of vowel sounds. This paper presents a magnetoencephalographic (MEG) experiment designed to locate and compare neuromagnetic activity associated with voice pitch, MFF and vowel type in human auditory cortex. Sequences of six sustained vowels were used to contrast changes in the three components of vowel perception, and MEG responses to the changes were recorded from 25 participants. A staged procedure was employed to fit the MEG data with a source model having one bilateral pair of dipoles for each component of vowel perception. This dipole model showed that the activity associated with the three perceptual changes was functionally separable; the pitch source was located in Heschl's gyrus (bilaterally), while the vowel-type and formant-frequency sources were located (bilaterally) just behind Heschl's gyrus in planum temporale. The results confirm that vowel normalization begins in auditory cortex at an early point in the hierarchy of speech processing.