Cantonese Tones Research Articles

Effects of tone training on Cantonese tone-word learning

The present study examined the effect of improving lexical tone identification abilities on Cantonese tone-word learning. Native English non-musicians received training on Cantonese tones before learning the meanings of words distinguished by these tones. Their results were compared to English non-musicians and musicians who received no tone training. The tone-trainees obtained a similar level of word identification proficiency as musicians by the end of training and were significantly better than non-tone trained non-musicians. These results lend support for phonetic-phonological-lexical continuity in learning because enhancing listeners' perception of lower-level tonal information significantly contributed to success in a higher-level linguistic task.

An ERP study of good production vis-à-vis poor perception of tones in Cantonese: implications for top-down speech processing.

This study investigated a theoretically challenging dissociation between good production and poor perception of tones among neurologically unimpaired native speakers of Cantonese. The dissociation is referred to as the near-merger phenomenon in sociolinguistic studies of sound change. In a passive oddball paradigm, lexical and nonlexical syllables of the T1/T6 and T4/T6 contrasts were presented to elicit the mismatch negativity (MMN) and P3a from two groups of participants, those who could produce and distinguish all tones in the language (Control) and those who could produce all tones but specifically failed to distinguish between T4 and T6 in perception (Dissociation). The presence of MMN to T1/T6 and null response to T4/T6 of lexical syllables in the dissociation group confirmed the near-merger phenomenon. The observation that the control participants exhibited a statistically reliable MMN to lexical syllables of T1/T6, weaker responses to nonlexical syllables of T1/T6 and lexical syllables of T4/T6, and finally null response to nonlexical syllables of T4/T6, suggests the involvement of top-down processing in speech perception. Furthermore, the stronger P3a response of the control group, compared with the dissociation group in the same experimental conditions, may be taken to indicate higher cognitive capability in attention switching, auditory attention or memory in the control participants. This cognitive difference, together with our speculation that constant top-down predictions without complete bottom-up analysis of acoustic signals in speech recognition may reduce one’s sensitivity to small acoustic contrasts, account for the occurrence of dissociation in some individuals but not others.

When does native language input affect phonetic perception? The precocious case of lexical tone

Previous studies have suggested that the perception of vowels and consonants changes from language-universal to language-specific between 6 and 12months of age. This report suggests that language-specific perception emerges even earlier for lexical tones. Experiment 1 tested English-learners’ perception of Cantonese tones, replicating declines in tone discrimination from 4 to 9months of age. Experiment 2 tested infants learning non-native versus native tone systems (Mandarin-learners versus Cantonese-learners). All Chinese-learners discriminated the tones, but showed language-specific differences in tone preferences at both ages. Indeed, English-, Mandarin-, and Cantonese-learning 4-month-olds all exhibited distinct preferences. With other work, this shows that language-specific speech perception emerges over a more complex and extended schedule than previously thought: first for lexical stress and tone (<5months), then vowels (6–8months), consonants (8.5–12months), and finally phoneme duration (18months). Acoustic salience likely plays an important role in determining the timing of phonetic development.

The separation between music and speech: Evidence from the perception of Cantonese tones

This study investigates the relationship between music and speech, testing whether musical training has any facilitatory effects on native tone language speakers. Some Cantonese tone pairs are merging in recent years. The merging subjects have poorer general lexical tone perception than the control subjects. Previous studies showed that musical training facilitates lexical tone perception of nontone language speakers, but it is unclear if the same is true for tone language speakers. Three groups of listeners (standard Cantonese, merging Cantonese, nontone) with and without musical training participated in AX discrimination tasks of Cantonese monosyllables and pure tones resynthesized from Cantonese lexical tones. Results show that while musical training enhances lexical tone perception of nontone listeners, it has little influence on Cantonese listeners. The findings suggest that the linguistic use of tones is more fundamental and more robust than musical tones. Our results are compatible with the idea that linguistic and musical mechanisms belong to separate but overlapping domains.

The Representation of Contour Tones in Cantonese

&lt;p&gt;A central question in tonal phonology is the representation of tone. One of the focal points is the representation of contour tones, especially since Goldsmith (1976) and subsequent works have analyzed contour tones in Bantu languages as sequences of level tones. Cross-linguistically, it is generally well-recognized, following Yip’s (1989) terminology, that contour tones in African languages are typically clusters, which are sequences of level tones and consist of multiple tonal root nodes, and that contour tones in Asian languages are typically tone units, which have only one tonal root node. This paper points out an important exception—Cantonese— particularly in light of Yip (2001) and Barrie (2007) on Chinese contour tones. The correct view is at least implied in earlier analyses: Cantonese tones, contour and level alike, should be represented as sequences of level tones but not unitary tone units.&lt;/p&gt;

Cross-language assimilation of lexical tone

We extend to lexical-tone systems a model of second-language perception, the Perceptual Assimilation Model (PAM) (Best &amp; Tyler, 2007), to examine whether/how native-language lexical-tone inventory composition influences perception of novel tone. Native listeners of Cantonese, Thai, and Mandarin perform a tone mapping-rating assimilation task. Listeners hear CV syllables bearing all tones of Cantonese, Thai, Mandarin, and Yoruba - languages with different tone inventories. They (1) map the tone they hear to the nearest native tone category, and (2) provide a goodness rating on a 5-point scale (5 = perfect). As predicted by the PAM, listeners assimilated non-native tones to the phonetically-closest native tone categories. Listeners attended primarily to pitch-contour, and secondarily to pitch-height, contrasts for the mappings. E.g., Mandarin listeners assimilated the Thai high “level” (phonetically mid-to-high-rising) tone to Mandarin rising tone 76% of the time, and to Mandarin high-level tone only 22% of the time. Also as predicted, all novel tones did not assimilate equally well to native categories; mappings received ratings between 2.9-4.1, averaging 3.5. The groups’ different patterns of results indicate that novel-tone perception is influenced by experience with the native-language tone inventory, and that listeners attend to gradient phonetic detail to assimilate novel tones to native-tone categories. This work is supported by NSF grant 0965227 to J.A.

The influence of linguistic and musical experience on Cantonese word learning

Adult non-native speech perception is subject to influence from multiple factors, including linguistic and extralinguistic experience such as musical training. The present research examines how linguistic and musical factors influence non-native word identification and lexical tone perception. Groups of native tone language (Thai) and non-tone language listeners (English), each subdivided into musician and non-musician groups, engaged in Cantonese tone word training. Participants learned to identify words minimally distinguished by five Cantonese tones during training, also completing musical aptitude and phonemic tone identification tasks. First, the findings suggest that either musical experience or a tone language background leads to significantly better non-native word learning proficiency, as compared to those with neither musical training nor tone language experience. Moreover, the combination of tone language and musical experience did not provide an additional advantage for Thai musicians above and beyond either experience alone. Musicianship was found to be more advantageous than a tone language background for tone identification. Finally, tone identification and musical aptitude scores were significantly correlated with word learning success for English but not Thai listeners. These findings point to a dynamic influence of musical and linguistic experience, both at the tone dentification level and at the word learning stage.

Relative contributions of temporal and spectral cues for Mandarin and Cantonese tone recognition

Despite good speech perception performance in quiet, pitch perception remains a challenge for cochlear-implant users. The current vocoder-based processing technique in cochlear implants preserves temporal envelope and coarse spectral information. This information, however, is insufficient to support high-level performance for tone perception in tonal languages (e.g., Mandarin and Cantonese), particularly in noise. Mandarin has four lexical tones and Cantonese has six lexical tones. Each lexical tone has different fundamental frequency contours. The temporal envelope of the signal also differs among different tones. In this talk, we will first discuss the relative contributions of temporal and spectral envelope and fine structure cues for Mandarin and Cantonese tone perception. We will then describe a new signal processing algorithm developed at the Chinese University of Hong Kong that aims to enhance temporal periodicity cues for Cantonese tone perception in noise. We will present findings from normal-hearing and cochlear-implant listeners, which demonstrate a significant improvement of Cantonese tone perception in noise with this new algorithm.

Extrinsic context is crucial for talker normalization in Cantonese tone perception

Previous studies showed that recognizing a phonetic category produced by different talkers relies on both intrinsic (target-internal) and extrinsic (contextual) cues. Extrinsic cues influence perception when intrinsic cues allow more than one phonetic interpretation. A recent study in this laboratory found that the configuration of tone systems (Cantonese and Mandarin) affects the degree of ambiguity of tones associated with intrinsic cues. In Cantonese which has three level tones, an isolated level pitch can be mapped to any of these three categories. Cantonese but not Mandarin listeners were found to confuse tones in a way biased by relative pitch height of different talkers. The present study tested Cantonese listeners on stimuli from four talkers with different pitch ranges (Female High, Female Low, Male High, and Male Low). Syllable /i/ carrying different F0 contours was embedded in a meaningful sentence with cues of a talker's F0 range. This study found enhanced identification accuracy with contextual cues over performance in isolation (92.25% vs. 51.75%), suggesting that extrinsic context facilitates talker normalization. This finding implies that extrinsic cues are especially useful for a language with intrinsically ambiguous phonetic categories. [Research supported by GRF 455911, NSFC 11074267, NSFC 61135003, and a 973 grant 2012CB720700.]

The Effect of Intertalker Variations on Acoustic–Perceptual Mapping in Cantonese and Mandarin Tone Systems

This study investigates the impact of intertalker variations on the process of mapping acoustic variations on tone categories in two different tone languages. Pitch stimuli manipulated from four voice ranges were presented in isolation through a blocked-talker design. Listeners were instructed to identify the stimuli that they heard as lexical tones in their native language. Tone identification of Mandarin listeners exhibited relatively stable normalization regardless of the voice, whereas tone identification of Cantonese listeners was unstable and susceptible to the influence of intertalker variations. In the case of Cantonese listeners, intertalker variations had a larger effect on the perception of F0 height dimension than of F0 slope dimension. The comparison between Cantonese and Mandarin listeners' performances reveals an interaction of intertalker variations and the types of tone contrasts in each language. For Cantonese tones, which depend heavily on F0 height distinctions, intertalker variations result in F0 overlapping and, consequently, ambiguities among them in isolated tone perception. For Mandarin tones, which are distinctive in terms of their F0 contours, the differences in F0 contours alone seem sufficient to elicit reliable tone identification. Intertalker variations therefore have relatively limited effect on Mandarin tone perception.

Effect of Tones on Vocal Attack Time in Cantonese Speakers

Vocal attack time (VAT) is the time lag between the growth of the sound pressure signal and the development of physical contact of vocal folds at vocal initiation. It can be derived by a cross-correlation of short-time amplitude changes occurring in the sound pressure and electroglottographic (EGG) signals. Cantonese is a tone language in which tone determines the lexical meaning of the syllable. Such linguistic function of tone has implications for the physiology of tone production. The aim of the present study was to investigate the possible effects of Cantonese tones on VAT. Sound pressure and EGG signals were simultaneously recorded from 59 native Cantonese speakers (31 females and 28 males). The subjects were asked to read aloud 12 disyllabic words comprising homophone pairs of the six Cantonese lexical tones. Results revealed a gender difference in VAT values, with the mean VAT significantly smaller in females than in males. There was also a significant difference in VAT values between the two tone categories, with the mean VAT values of the three level tones (tone 1, 3, and 6) significantly smaller than those of the three contour tones (tone 2, 4, and 5). The findings support the notion that norms and interpretations based on nontone European languages may not be directly applied to tone languages.

ERP correlates of pre-attentive processing of Cantonese lexical tones: The effects of pitch contour and pitch height

The pre-attentive processing of Cantonese tones was studied with an auditory passive oddball paradigm. Event-related potentials to standard and deviant auditory stimuli were recorded as participants watched a silent movie attentively. The standards and deviants differed in either pitch level or pitch contour. Mismatch negativity (MMN) and P3a were elicited by all types of deviant tones, suggesting that lexical tone was processed pre-attentively. In addition, the size and latency of MMN were sensitive to the size of pitch level change, while the latency of P3a captured the presence of pitch contour change. These results indicate that pitch contour and pitch height are two important dimensions in sensory processing of lexical tones.

The influence of language experience on categorical perception of pitch contours

Previous research on categorical perception of pitch contours has mainly considered the contrast between tone language and non-tone language listeners. This study investigates not only the influence of tone language vs. non-tone language experience (German vs. Chinese), but also the influence of different tone inventories (Mandarin tones vs. Cantonese tones), on the categorical perception of pitch contours. The results show that the positions of the identification boundaries do not differ significantly across the 3 groups of listeners, i.e., Mandarin, Cantonese, and German, but that the boundary widths do differ significantly between tone language (Mandarin and Cantonese) listeners and non-tone language (German) listeners, with broader boundary widths for non-tone language listeners. In the discrimination tasks, the German listeners exhibit only psychophysical boundaries, whereas Chinese listeners exhibit linguistic boundaries, and these linguistic boundaries are further shaped by the different tone inventories.

Can musical aptitude and experience predict success in non‐native tone word learning?

Non‐native linguistic pitch perception is subject to influence from a variety of factors in addition to linguistic background, including musical experience. The present study investigated the effects of musical aptitude and musical experience on Cantonese tone word learning and how these musical factors interact with linguistic experience. Adult native Thai listeners whose native language (L1) is tonal and English listeners with a non‐tonal L1, subdivided into musician and non‐musician groups, engaged in a seven‐session perceptual training program, learning the meanings of 15 novel vocabulary words distinguished by five Cantonese lexical tones. Before training, a musical aptitude task was administered to establish the participants’ level of musicality and auditory discrimination skills. The results show significant group differences in tone word learning proficiency by the end of training. English musicians outperformed their non‐musician counterparts; however, the Thai musicians were not significantly different from the Thai non‐musicians. Results from regression analyzes further show that higher musical aptitude scores predicted tone word learning success for the English group but not for the Thai group. These findings suggest that the influence of musical experience in constructing novel lexical representations of tone words differs as a function of linguistic background.

Lexical Tone Perception Ability of Profoundly Hearing-Impaired Children

In tone languages such as Cantonese, a change in tone denotes a change in lexical meaning. The present study investigates the functional benefit of hearing devices in assisting tone perception among children with profound hearing impairment. Fifty-two children with profound hearing loss were categorized into two groups based on their primary type of hearing device - a hearing aid group and cochlear implant group. A 75-item tone identification test covering all 15 Cantonese tone contrast pairs was administered to each subject under two conditions - unaided (hearing devices turned off) and aided (devices turned on). The proportion of correct responses was computed as the total score for all items and subtotal contrast scores for each of the 15 tone contrasts. The results indicated no significant differences between the children wearing hearing aids and those with cochlear implants under the unaided or the aided condition (z = -0.91, p = 0.36; z = -0.40, p = 0.69, respectively). Regardless of the type of device used, the total scores under the aided condition were higher than those under the unaided condition (z = -3.55 for the hearing aid group; z = -4.87 for the cochlear implant group, both ps < 0.01). Children wearing hearing devices generally have a satisfactory functional gain in tone perception. No major observable difference was noted between children using cochlear implants and those using hearing aids. Tone contrast pairs with dissimilar fundamental frequency at onset and dissimilar tone contour patterns were more easily identified.

Age Sensitivity in the Acquisition of Lexical Tone Production: Evidence from Children with Profound Congenital Hearing Impairment after Cochlear Implantation

This study investigated the effects of implant experience and age at implantation on the Cantonese tone production of children with cochlear implants. The study also examined whether there was a particular age at which children were more responsive to acquiring tones. The study included 45 children who had received unilateral cochlear implants at a mean age of 65.56 months. The subjects were grouped according to their age at cochlear implantation and were assessed annually for 5 years thereafter. A picture-naming task was used to measure their tone production performance. A simple effect of age at implantation was significant at all testing intervals except at the preoperative data point. Children who were younger than 4 years of age when they received their implants scored significantly higher than did the 2 older groups at various testing intervals. A significant simple effect of implant experience was also found. Progress was most striking in children who received their implants before the age of 4 years. For the most effective acquisition of Cantonese lexical tones, children should undergo early cochlear implantation. For children who receive implants before the age of 4 years, benefits are noted in tone production ability in terms of a faster rate of improvement within a shorter period of time.

An acoustic and electroglottographic study of Cantonese tone.

Cantonese is a tone language with six tones (traditionally described as high level, high rising, mid level, mid-low falling, mid-low rising, and mid-low level), and also three “stopped” tones in CVC words, not studied here. The interest of this study was (1) how different acoustic cues can be used to classify the different tones, particularly in connected speech and (2) the interaction of voice quality and tone in Cantonese. Twelve speakers were recorded producing minimal pairs as in Wong [(2006)] in isolation, in isolated disyllables, and in sentence-medial disyllables, where the disyllables ranged over all possible 36 bitones in the language; all speakers also made electroglottographic (EGG) recordings. Acoustic measures were F0, F0′, cepstral peak prominence, and harmonic amplitudes H1* and H2*, H1*-H2*, H1*-A1*, H1*-A2*, H1*-A3*, and H2*-H4*. EGG measures were closed quotient and peak of increasing contact. Measures were made automatically using VOICESAUCE and EGGWORKS. Preliminary results indicate variability in the presence of creak in the realization of the lowest tone (mid-low falling) as well as variability in the presence of creak at the low f0 regions of the rises.

A Study of the Cantonese Hearsay Particle wo from a Tonal Perspective

Cantonese is a tone language very rich in sentence-final particles which express moods and attitudes. Some of them are identical in phonetic segments but different in lexical tones. This paper discusses a pair of such SFPs in Cantonese, wo3 and wo5, the former has been in use for a long time and its function has altered from indicating hearsay to unexpectedness and noteworthiness, while the latter is not found in the earlier literature and its only function is to express hearsay information. By reviewing the development of Chinese phonology and present-day Cantonese phonology, hypotheses are proposed in this paper to explain the reason why the tone of the hearsay wo would be tone 5 instead of the other tones when there are six contrastive lexical tones in Hong Kong Cantonese.

Cantonese tone recognition with enhanced temporal periodicity cues

This study investigated the contributions of temporal periodicity cues and the effectiveness of enhancing these cues for Cantonese tone recognition in noise. A multichannel noise-excited vocoder was used to simulate speech processing in cochlear implants. Ten normal-hearing listeners were tested. Temporal envelope and periodicity cues (TEPCs) below 500 Hz were extracted from four frequency bands: 60-500, 500-1000, 1000-2000, and 2000-4000 Hz. The test stimuli were obtained by combining TEPC-modulated noise signals from individual bands. For periodicity enhancement, temporal fluctuations in the range 20-500 Hz were replaced by a sinusoid with frequency equal to the fundamental frequency of original speech. Tone identification experiments were carried out using disyllabic word carriers. Results showed that TEPCs from the two high-frequency bands were more important for tone identification than TEPCs from the low-frequency bands. The use of periodicity-enhanced TEPCs led to consistent improvement of tone identification accuracy. The improvement was more significant at low signal-to-noise ratios, and more noticeable for female than for male voices. Analysis of error distributions showed that the enhancement method reduced tone identification errors and did not show any negative effect on the recognition of segmental structures.

Tonal confusion patterns in Mandarin by Cantonese listeners.

It is well‐attested that linguistic experience affects the perception of non‐native sounds. The vast majority of research on L2 perception has been carried out on segmental features from the perspective of phonetic similarity between the L1 and L2 sound systems. The general goal of this study is to expand our understanding of cross‐linguistic comparison in non‐native speech perception, focusing on the somewhat understudied area of perceptual assimilation at a suprasegmental level. More specifically, it deals with comparisons of prosodic systems in two tone languages, Mandarin and Cantonese. The current study examined perception of Mandarin tones by Cantonese listeners. In Experiment 1, Mandarin tones were presented and the Cantonese listeners were requested to identify which tone they heard. In Experiment 2, the Cantonese listeners were instructed to rate how similar each Mandarin tone was to a Cantonese tone. Preliminary results suggest that tonal confusion errors may result from not only the similar acoustic properties of the tone pairs but also perceptual assimilation between L1 and L2 tonal contrasts. These findings are discussed in terms of the effects of L1 prosodic system on L2 perception and how perceptually assimilation patterns predict listeners’ perception performance at the domain of lexical tones.