Cantonese Tones Research Articles

The Effect of Speech Variability on Tonal Language Speakers' Second Language Lexical Tone Learning.

Speech variability facilitates non-tonal language speakers’ lexical tone learning. However, it remains unknown whether tonal language speakers can also benefit from speech variability while learning second language (L2) lexical tones. Researchers also reported that the effectiveness of speech variability was only shown on learning new items. Considering that the first language (L1) and L2 probably share similar tonal categories, the present study hypothesizes that speech variability only promotes the tonal language speakers’ acquisition of L2 tones that are different from the tones in their L1. To test this hypothesis, the present study trained native Mandarin (a tonal language) speakers to learn Cantonese tones with either high variability (HV) or low variability (LV) speech materials, and then compared their learning performance. The results partially supported this hypothesis: only Mandarin subjects’ productions of Cantonese low level and mid level tones benefited from the speech variability. They probably relied on the mental representations in L1 to learn the Cantonese tones that had similar Mandarin counterparts. This learning strategy limited the impact of speech variability. Furthermore, the results also revealed a discrepancy between L2 perception and production. The perception improvement may not necessarily lead to an improvement in production.

Cantonese Tone Identification in Three Temporal Cues in Quiet, Speech-Shaped Noise and Two-Talker Babble

Purpose: Cochlear implant processors deliver mostly temporal envelope information and limited fundamental frequency (F0) information to the users, which make pitch and lexical tone perception challenging for cochlear implantees. Different factors have been found to affect Mandarin tone perception in temporal cues but the most effective temporal cues for lexical tone identification across different backgrounds remained unclear because no study has comprehensively examined the effects and interactions of these factors, particularly, in languages that use both pitch heights and pitch shapes to differentiate lexical meanings. The present study compared identification of Cantonese tones in naturally produced stimuli, and in three temporal cues, namely the amplitude contour cue (TE50), the periodicity cue (TE500), and the temporal fine structure cue (TFS), in three different numbers of frequency bands (B04, B08, B16) in quiet and two types of noise (two male talker-babble and speech-shaped noise).Method: Naturally produced Cantonese tones and synthetic tones that combined different acoustic cues and different number of frequency bands were presented to 18 young native Cantonese speakers for tone identification in quiet and noise.Results: Among the three temporal cues, TFS was the most effective for Cantonese tone identification in quiet and noise, except for T4 (LF) identification. Its effect was even stronger when the tones were presented in 4 or 8 bands rather than 16 bands. Neither TE500 nor TE50 was effective for Cantonese tone identification in quiet or noise. In noise, most tones in TE500 and TE50 were misheard as T4 (LF), demonstrating errors in both tone shapes and tone heights. Types of noise had limited effect on tone identification.Conclusions: Findings on Mandarin tone perception in temporal cues may not be applicable to other tone languages with more complex tonal systems. TFS presented in four bands was the most effective temporal cue for Cantonese tone identification in quiet and noise. Temporal envelope cues were not effective for tone, tone shape or tone height identification in Cantonese. These findings have implications for future design of cochlear implants for tone speakers who use pitch heights or a combination of pitch heights and pitch shapes to differentiate meanings.

Acoustic-phonetic consequences of clear speech in Cantonese vowels and lexical tones

In adverse communicative contexts, speakers will modify their speech style to enhance intelligibility, which can involve a range of acoustic and articulatory adjustments. Prior work has focused on the influence of speaking style on spectral (e.g., vowel space expansion) and temporal distinctions (e.g., voice-onset time). Thus, this study investigated the extent to which such adjustments extend to the pitch dimension, namely lexical tone. Ten native Cantonese speakers were recorded producing five monosyllables with six Cantonese tones embedded in sentence contexts in both clear and conversational speaking styles. Acoustic analyses revealed that speakers produced two distinct styles, with longer vowel durations and greater vowel dispersion in clear relative to conversational speech. However, with regards to lexical tone production, the results indicated only a marginal re-adjustment of tones within the tone space as a result of clear speech. Specific contour tones, such as the low- and high-rising tones, saw increases in their F0 offset. However, mean F0, F0 range and tonal dispersion did not differ substantially between styles. These findings indicate that speakers maintain tonal stability across speaking styles, suggesting that, relative to segments, lexical tones may not be prioritized as phonological contrasts vital to enhance the overall intelligibility of the utterance.

Testing the Hyperarticulation and Prosodic Hypotheses of Child-Directed Speech: Insights From the Perceptual and Acoustic Characteristics of Child-Directed Cantonese Tones

The function of child-directed speech has been debated for decades. This study examined the perceptual and acoustic characteristics of child- and adult-directed Cantonese tones to test the hyperarticulation and prosodic hypotheses that have been proposed to account for the acoustic modifications in child-directed speech. Sixty-two mother-child dyads participated in the study. The mothers verbally labeled 30 pictures in monosyllabic isolated words and in the final position of a carrier sentence to the experimenter and their 1- to 5-year-old children. The 8,634 adult- and child-directed productions were low-pass filtered to eliminate lexical information and presented to 5 judges for tone identification. Acoustic analysis was performed on the productions. Acoustically, child-directed tones were produced with an elevated pitch, and the pitch level decreased as the child's age increased. Acoustic contrasts between phonetically similar and more confusing tones were not enhanced in child-directed speech, and unexpectedly, child-directed tones were identified with a lower accuracy than adult-directed tones. The perceptual errors of child-directed tones mirrored the errors found in identifying tones excised from sentence-final position, which had a pitch-lowering effect on the tones. The lower perceptual accuracy, the lack of enhanced acoustic contrasts in confusing tone pairs, and the similarities in the error patterns in identifying tones in child-directed speech and tones in utterance-final position suggest that the acoustic modifications in child-directed tones are prosodic effects serving pragmatic purposes. The findings reject the hyperarticulation hypothesis and support the prosodic hypothesis of child-directed speech.

The acquisition of lexical tones by Cantonese-English bilingual children.

Previous studies on bilingual children found intact tonal development at the initial stages of interaction between Cantonese and English in successive bilingual children, whereas children exposed to both languages from birth have not been studied in this regard. We examined the production of Cantonese tones by five simultaneous bilingual children longitudinally at 2;0 and 2;6, and compared them with age-matched monolingual children using auditory analysis. Our results showed that some bilingual children had a delay at 2;0, compared to their monolingual peers. Some bilingual children also exhibited a 'high-low' template in their production, resembling the pitch pattern of English trochaic words. These findings suggest a possible early interaction of the Cantonese and English prosodic systems in which bilingual children adopted the English stress pattern in Cantonese production. The time-point along the trajectory of phonological development is important in modulating whether cross-linguistic transfer can be observed.

Suprasegmental Features Are Not Acquired Early: Perception and Production of Monosyllabic Cantonese Lexical Tones in 4- to 6-Year-Old Preschool Children.

Previous studies reported that children acquire Cantonese tones before 3 years of age, supporting the assumption in models of phonological development that suprasegmental features are acquired rapidly and early in children. Yet, recent research found a large disparity in the age of Cantonese tone acquisition. This study investigated Cantonese tone development in 4- to 6-year-old children. Forty-eight 4- to 6-year-old Cantonese-speaking children and 28 mothers of the children labeled 30 pictures representing familiar words in the 6 tones in a picture-naming task and identified pictures representing words in different Cantonese tones in a picture-pointing task. To control for lexical biases in tone assessment, tone productions were low-pass filtered to eliminate lexical information. Five judges categorized the tones in filtered stimuli. Tone production accuracy, tone perception accuracy, and correlation between tone production and perception accuracy were examined. Children did not start to produce adultlike tones until 5 and 6 years of age. Four-year-olds produced none of the tones with adultlike accuracy. Five- and 6-year-olds attained adultlike productions in 2 (T5 and T6) to 3 (T4, T5, and T6) tones, respectively. Children made better progress in tone perception and achieved higher accuracy in perception than in production. However, children in all age groups perceived none of the tones as accurately as adults, except that T1 was perceived with adultlike accuracy by 6-year-olds. Only weak association was found between children's tone perception and production accuracy. Contradicting to the long-held assumption that children acquire lexical tone rapidly and early before the mastery of segmentals, this study found that 4- to 6-year-old children have not mastered the perception or production of the full set of Cantonese tones in familiar monosyllabic words. Larger development was found in children's tone perception than tone production. The higher tone perception accuracy but weak correlation between tone perception and production abilities in children suggested that tone perception accuracy is not sufficient for children's tone production accuracy. The findings have clinical and theoretical implications.

Perceptual Reorganization of Lexical Tones: Effects of Age and Experimental Procedure.

Findings on the perceptual reorganization of lexical tones are mixed. Some studies report good tone discrimination abilities for all tested age groups, others report decreased or enhanced discrimination with increasing age, and still others report U-shaped developmental curves. Since prior studies have used a wide range of contrasts and experimental procedures, it is unclear how specific task requirements interact with discrimination abilities at different ages. In the present work, we tested German and Cantonese adults on their discrimination of Cantonese lexical tones, as well as German-learning infants between 6 and 18 months of age on their discrimination of two specific Cantonese tones using two different types of experimental procedures. The adult experiment showed that German native speakers can discriminate between lexical tones, but native Cantonese speakers show significantly better performance. The results from German-learning infants suggest that 6- and 18-month-olds discriminate tones, while 9-month-olds do not, supporting a U-shaped developmental curve. Furthermore, our results revealed an effect of methodology, with good discrimination performance at 6 months after habituation but not after familiarization. These results support three main conclusions. First, habituation can be a more sensitive procedure for measuring infants' discrimination than familiarization. Second, the previous finding of a U-shaped curve in the discrimination of lexical tones is further supported. Third, discrimination abilities at 18 months appear to reflect mature perceptual sensitivity to lexical tones, since German adults also discriminated the lexical tones with high accuracy.

Recent findings on children’s lexical tone development: Implications for models of speech development

Lexical tone is important for a majority of the world’s languages. Prior research consistently reported that children mastered lexical tone production between 1;6 to 2;6 of age, supporting the prevailing notion of theories of speech acquisition that children acquire supra-segmental features early and rapidly before mastering the full set of segmental features. However, intriguingly, studies that examined children’s Cantonese tone perception found that children failed to accurately identify Cantonese tones in monosyllabic words until after eight years old, five years after children’s full mastery of tone production at three years old, which questioned early acquisition of supra-segmental features and the widely accepted assumption in speech acquisition models that speech perception precedes speech production. Wong, Schwartz, and Jenkins (2005) started a series of studies using a new paradigm that controlled lexical expectation in tone rating to examine lexical tone development in children and reported strikingly different results. Monosyllabic and disyllabic tones produced by five and six years old Mandarin-speaking children growing up in the U.S. and in Taiwan and Cantonese-speaking children growing up in Hong Kong were not adult-like. Tone production development was slower than tone perception development. The findings shed new light on universal models and theories of speech acquisition.

Acoustic characteristics of highly distinguishable Cantonese entering and non-entering tones.

Cantonese has one of the most complex tone systems. Few studies have thoroughly examined or compared the acoustic properties of the full set of Cantonese tones, particularly the entering tones, compromising deeper understanding of Cantonese tone difficulties in various clinical populations. This study (1) describes a theory-driven method for acoustic analysis of tones that successfully normalized the intrinsic pitch of male and female speakers, (2) provides detailed acoustic data on distinctly enunciated Cantonese tones, (3) examines the acoustic similarities and differences between the entering and non-entering tones, and (4) compares the acoustic properties of three easily confused tone pairs. Seventeen male and female native speakers produced 1802 Cantonese tones that were correctly identified by five judges in filtered stimuli. Counter to the established notion that the entering tones are shorter versions of the three level tones, the results revealed that the entering tones have falling contours, suggesting that the entering and non-entering tones should be examined separately in research and clinical settings. The detailed description of the acoustic properties of the nine tones and the acoustic contrasts of the entering and non-entering tones and the three easily confused tone pairs provides references for future Cantonese tone studies with different populations.

Perception of Cantonese Lexical Tones by Pediatric Cochlear Implant Users.

The purpose of this study is to assess Cantonese word recognition and the discrimination of Cantonese tones with manipulated contours by child and adolescent cochlear implant (CI) users and a group of peers with normal hearing (NH). It was hypothesized that the CI users would perform more poorly than their counterparts with NH in both tasks and that CI users implanted before 2 years of age would perform better than those implanted after 2 years. Forty-one participants were recruited from hospitals, schools, and kindergartens in Hong Kong: Ten CI users implanted at or before 2 years of age ("early" CI group), 13 CI users implanted after 2 years of age ("late" CI group), and 18 individuals with NH. The mean age at implantation of the early CI group was 1.5 years (SD = 0.3), and for the late CI group, it was 4.3 years (SD = 2.1). Participants were a mean of 13.3 years of age (SD = 3.7) at time of testing. Participants completed a Cantonese word recognition test and a discrimination task using Cantonese tones with modified fundamental frequency trajectories. Both CI user groups obtained significantly lower scores than the group with NH on the word recognition test. Mean percent correct scores for the word recognition test were 79% for the early CI group, 69% for the late CI group, and 97% for the group with NH. The group with NH consistently achieved higher scores than the CI user groups when discriminating manipulated Cantonese tones. Increasing the acoustic difference between tones improved discrimination performance for CI users for level tone contrasts only. CI users implanted at or before 2 years of age obtained higher scores than those implanted later. The results of this study add further evidence that children using CIs do not perform as well as peers in perceiving Cantonese tones. Modification of tones to increase pitch range did not consistently improve the ability of children with implants to perceive the difference between tones. Further research is required to fully assess potential benefits of early implantation for speakers of tonal languages. https://doi.org/10.23641/asha.5782209.

Native Hakka speakers′ identification of Cantonese tones: A comparative analysis

Objective To investigate the features of native Hakka speakers′ ability to identify Cantonese tones. Methods Twenty native Cantonese speakers and 12 native Hakka speakers were given the Hongkong Cantonese tone identification test. Results Neither group was troubled by similar vowel and consonant sounds. The Cantonese group could identify all tones well except distinguishing tones 5 and 2 (89%) and tones 4 and 6 (89%). The Hakkas had the most difficulty distinguishing tones 3 and 6 (70.0%), but significant differences were observed between the two groups in the identification of tones 2 and 5 and 2 and 4. Conclusion A native Hakka speaker′s ability to identify Cantonese tones may be closely related to the tone′s pitch, tonal patterns as well as the Hakka tone system. Key words: Cantonese; Hakka; Tones

Cantonese Tone Perception for Children Who Use a Hearing Aid and a Cochlear Implant in Opposite Ears.

The ability to recognize tones is vital for speech perception in tonal languages. Cantonese has six tones, which are differentiated almost exclusively by pitch cues (tones 1 to 6). The differences in pitch contours among the tones are subtle, making Cantonese a challenging language for cochlear implant users. The addition of a hearing aid has been shown to improve speech perception in nontonal languages and in Mandarin Chinese. This study (1) investigates the Cantonese tone perception ability of children who use a cochlear implant and a hearing aid in opposite ears; (2) evaluates the effect of varying pitch height and pitch contour cues on Cantonese tone perception for these children; and (3) compares the Cantonese tone perception ability for using a hearing aid and a cochlear implant together versus an implant alone. Eight native Cantonese speaking children using a cochlear implant and a hearing aid in opposite ears were assessed for tone perception and word identification. The tone perception test involved discriminating and ranking tone pairs from natural and artificially manipulated Cantonese tones with various pitch heights and/or pitch contours. The word identification test involved identifying Cantonese words in a four-alternative forced-choice task. All tests were performed in two device conditions: (1) cochlear implant and hearing aid together and (2) implant alone. Seven of the 8 subjects performed significantly above chance in both tests using the cochlear implant alone. Results showed that both pitch height and/or pitch direction were important perceptual cues for implant users. Perception for some tones was improved by increasing the pitch height differences between the tones. The ability to discriminate and rank the tone 2/tone 5 contrast and the tone 4/tone 6 contrast was poor, as the tones in these contrasts are similar in pitch contours and onset frequencies. No significant improvement was observed after artificially increasing the pitch offset differences between the tones in the tone 2/tone 5 and the tone 4/tone 6 contrasts. Tone perception results were significantly better with the addition of the hearing aid in the nonimplanted ear compared with using the implant alone; however, word identification results were not significantly different between using the implant alone and using both the hearing aid and the implant together. None of the subjects performed worse in tone perception or in word identification when the hearing aid was added. Reduced ability to perceive pitch contour cues, even when artificially exaggerated, may explain some of the difficulties in Cantonese word recognition for implant users. The addition of a contralateral hearing aid could be beneficial for Cantonese tone perception for some individuals with a unilateral implant. The results encouraged Cantonese speakers to trial a hearing aid in the nonimplanted ear when using a cochlear implant.

Native perception of Cantonese tones in high-variability conditions

This study investigates how native speakers of Cantonese perceive and produce Cantonese tone pairs T2-T5, T3-T6, T4-T5, T4-T6, and T5-T6 in high-variability conditions. Thirty five native speakers of Cantonese from Hong Kong participated in a 6AFC identification task (screening test), a mixed-talker AXB task (ISI = 1500 ms), and a mixed-talker repetition task (delay = 1500 ms). Results from the screening test showed that the participants did not merge the tones examined in this study. Results from the AXB task showed that (1) perception of tones with f0-height differences (T2-T5 and T3-T6) was more challenging than perception of tones with f0-contour differences (T4-T5, T4-T6, and T5-T6), (2) tones contrasting in direction of f0 change (T4-T5) were the easiest, and (3) perception of level tones (T3-T6) was more difficult than perception of contour tones with an f0-height difference (T2-T5). Data from the repetition task were fitted with quadratic polynomial equations (y = a + bx + cx2) and the analysis revealed that the participants produced the two level tones—T3 and T6—distinctively (in terms of coefficient-a) and the two rising tones—T2 and T5—distinctively (in terms of coefficient-c). This indicates that the participants perceived different level tones and different rising tones in the repetition task (i.e., no tone merger).

Lexical tone and stuttering in Cantonese

ABSTRACTCantonese is a tone language, in which the variation of the fundamental frequency contour of a syllable can change meaning. There are six different lexical tones in Cantonese. While research with Western languages has shown an association between stuttering and syllabic stress, nothing is known about whether stuttering in Cantonese speakers is associated with one or more of the six lexical tones. Such an association has been reported in conversational speech in Mandarin, which is also a tone language, but which varies markedly from Cantonese. Twenty-four native Cantonese-speaking adults who stutter participated in this study, ranging in age from 18–33 years. There were 18 men and 6 women. Participants read aloud 13 Cantonese syllables, each of which was produced with six contrastive lexical tones. All 78 syllables were embedded in the same carrier sentence, to reduce the influence of suprasegmental or linguistic stress, and were presented in random order. No significant differences were found for stuttering moments across the six lexical tones. It is suggested that this is because lexical tones, at least in Cantonese, do not place the task demands on the speech motor system that typify varying syllabic stress in Western languages: variations not only in fundamental frequency, but also in duration and intensity. The findings of this study suggest that treatments for adults who stutter in Western languages, such as speech restructuring, can be used with Cantonese speakers without undue attention to lexical tone.

Cantonese-Speaking Children Do Not Acquire Tone Perception before Tone Production-A Perceptual and Acoustic Study of Three-Year-Olds' Monosyllabic Tones.

Models of phonological development assume that speech perception precedes speech production and that children acquire suprasegmental features earlier than segmental features. Studies of Chinese-speaking children challenge these assumptions. For example, Chinese-speaking children can produce tones before two-and-a-half years but are not able to discriminate the same tones until after 6 years of age. This study compared the perception and production of monosyllabic Cantonese tones directly in 3 -year-old children. Twenty children and their mothers identified Cantonese tones in a picture identification test and produced monosyllabic tones in a picture labeling task. To control for lexical biases on tone ratings, the mother- and child-productions were low-pass filtered to eliminate lexical information and were presented to five judges for tone classification. Detailed acoustic analysis was performed. Contrary to the view that children master lexical tones earlier than segmental phonemes, results showed that 3-year-old children could not perceive or produce any Cantonese tone with adult-like proficiency and incorrect tone productions were acoustically different from criterion. In contrast to previous findings that Cantonese-speaking children mastered tone production before tone perception, we observed more accuracy during speech perception than production. Findings from Cantonese-speaking children challenge some of the established tenets in theories of phonological development that have been tested mostly with native English speakers.

The role of time in phonetic spaces: Temporal resolution in Cantonese tone perception

The role of temporal resolution in speech perception (e.g. whether tones are parameterized with fundamental frequency sampled every 10ms, or just twice in the syllable) is sometimes overlooked, and the temporal resolution relevant for tonal perception is still an open question. The choice of temporal resolution matters because how we understand the recognition, dispersion, and learning of phonetic categories is entirely predicated on what parameters we use to define the phonetic space that they lie in. Here, we present a tonal perception experiment in Cantonese where we used interrupted speech in trisyllabic stimuli to study the effect of temporal resolution on human tonal identification. We also performed acoustic classification of the stimuli with support vector machines. Our results show that just a few samples per syllable are enough for humans and machines to classify Cantonese tones with reasonable accuracy, without much difference in performance from having the full speech signal available. The confusion patterns and machine classification results suggest that loss of detailed information about the temporal alignment and shape of fundamental frequency contours was a major cause of decreasing accuracy as resolution decreased. Moreover, machine classification experiments show that for accurate identification of rising tones in Cantonese, it is crucial to extend the temporal window for sampling to the following syllable, due to peak delay.

Do mothers enhance the tonal contrasts in their monosyllabic Cantonese tones directed to their infants?

Introduction Some studies reported that adults enhanced acoustic differences of phonemes when speaking to young children. We examined whether mothers improved the contrasts among the lexical tones when speaking to infants. Methods Nineteen native Cantonese-speaking mothers produced the six Cantonese tones in 45 monosyllabic words to an adult and their 7- to 12-month-old infants. The 1709 words were low-pass filtered to preserve the pitch contours but eliminate lexical information. Five judges categorized the tones of the filtered words. Acoustic analysis was performed. Results Infant-directed tones had higher fundamental frequency (F0) and longer duration, but were identified with lower, though not significantly different, accuracy. Larger mean F0 differences were found in four pair of tones in infant-directed speech. However, these increased acoustic contrasts did not lead to higher perceptual accuracy of these tone pairs. Despite substantial perceptual confusion between T2 (HR) and T5 (LR) has been reported in previous tone studies, the difference in the slopes of the two tones was not enhanced in infant-directed speech. Conclusion Mothers acoustically modified their tones when speaking to infants. However, little evidence supported that mothers enhanced the phonetic contrasts of the tones in infant-directed speech. [Work supported by Research Grants Council of Hong Kong.]

The interaction between phonological information and pitch type at pre-attentive stage: an ERP study of lexical tones

ABSTRACTPrevious studies on the processing of lexical tones have typically confounded effects due to phonological information (different meanings of words signalled by syllables with different tonal categories) with effects due to specific acoustic information (pitch type: pitch height/pitch contour). The present study is designed to dissociate these two kinds of effects and further investigate the processing of lexical tones at pre-attentive stage by mismatch negativity (MMN). We chose level tones and contour tones in Cantonese to differentiate pitch height from pitch contour, and manipulated tonal category (within-category/across-category) to distinguish phonological information from acoustic information. The results showed clear interactions between tonal category and pitch type in MMN mean amplitude and peak latency, suggesting the interaction between phonological information and pitch type in the pre-attentive processing of lexical tones. These results are discussed in light of cognitive and neural mechanisms underlying auditory processing of lexical tones.

Applying Rasch model analysis in the development of the cantonese tone identification test (CANTIT)

Objective: Applying Rasch analysis to evaluate the internal structure of a lexical tone perception test known as the Cantonese Tone Identification Test (CANTIT). Design: A 75-item pool (CANTIT-75) with pictures and sound tracks was developed. Respondents were required to make a four-alternative forced choice on each item. A short version of 30 items (CANTIT-30) was developed based on fit statistics, difficulty estimates, and content evaluation. Internal structure was evaluated by fit statistics and Rasch Factor Analysis (RFA). Study Sample: 200 children with normal hearing and 141 children with hearing impairment were recruited. Results: For CANTIT-75, all infit and 97% of outfit values were < 2.0. RFA revealed 40.1% of total variance was explained by the Rasch measure. The first residual component explained 2.5% of total variance in an eigenvalue of 3.1. For CANTIT-30, all infit and outfit values were < 2.0. The Rasch measure explained 38.8% of total variance, the first residual component explained 3.9% of total variance in an eigenvalue of 1.9. Conclusions: The Rasch model provides excellent guidance for the development of short forms. Both CANTIT-75 and CANTIT-30 possess satisfactory internal structure as a construct validity evidence in measuring the lexical tone identification ability of the Cantonese speakers.

Verbal short-term memory and language impairments in Cantonese speakers after stroke

Purpose: The study examined the relationship between verbal short-term memory (STM) and language impairment in Cantonese speakers after stroke. It is hypothesised that Cantonese speakers with left-hemisphere (LH) stroke would perform worse than those with right hemisphere (RH) stroke and normal controls. Specific linguistic factors of Cantonese might affect results in the tasks.Method: Fifteen participants with LH stroke, 10 with RH stroke and 25 healthy controls were tested with auditory–verbal immediate serial recall (ISR) tasks and auditory linguistic tasks. All stroke participants were assessed with the Cantonese version of Western Aphasia Battery (CAB).Result: The LH group performed significantly worse than the RH and healthy control groups in the auditory verbal ISR and auditory linguistic tasks. There were significant lexicality, frequency and imageability effects in most tasks. Auditory discrimination and word comprehension tasks, but not the auditory word recognition task had correlations with ISR tasks.Conclusion: Verbal STM and language performance of Cantonese-speakers with history of LH stroke were inferior to RH stroke and healthy controls. The effects of lexicality, word frequency and imageability on verbal STM memory performance were found. Cantonese tones have effects on performance in auditory word recognition task, similar to onset, nucleus and rime.