Context Matters for Tone and Intonation Processing in Mandarin

Speech Prosody 2016, 2016

This study investigated how Mandarin listeners process tone and intonation when the F 0 encodings of the lexical tone and intonation are in conflict or in congruency and the role context plays during these processes. Tone and intonation identification experiments were conducted within neutral vs. constraining semantic contexts. Tone identification was much easier than intonation identification irrespective of contexts. Participants could perceive tones accurately and quickly in both question and statement intonation. However, intonation identification was greatly deteriorated within the neutral semantic context. Questions ending with a rising tone and a falling tone were equally difficult to identify. In a constraining semantic context, questions ending with a falling tone were much better identified. Thus, top-down information provided by the constraining semantic context does play an important role in disentangling intonation information from tone information.

International Journal on Studies in English Language and Literature, 2016

How does intonation affect lexical tone in Mandarin? This study examines the interaction of intonation and lexical tone in Mandarin. The study examines specifically the effect of intonational pitch on Tone 2 and Tone 4 in Mandarin. For the purpose of this experiment, we have examined a set of sentences uttered by three Mandarin speakers. The sentences have been uttered using both Tone 2 and Tone 4. Each of these tones has been used in both questions and statements. The study claims that there are more than two linguistic uses for F0 (fundamental frequency): lexical tone showing contrasts between words, and intonation distinguishing, among other functions, between different semantic and/or grammatical interpretations of two phrases. It also claims that there is a kind of simultaneous lexical and intonational pitch requirements on a single word. The study concludes that F0 at the end of yes/no questions is higher than F0 at the end of statements in Mandarin. It also concludes that there is some effect of tone on each of the intonation contour although the t-test indicates that the difference is not significant

1982

2004

Do lexical tones interfere with the realization of intonation types? Given that tone and intonation both use F0 as a primary cue, can a listener reliably identify statements and questions when some of the channel capacity is taken up by lexical tones? We study this issue through a perception test on a carefully designed and obtained intonation corpus on Mandarin Chinese. Our study shows the following: 1. Statement intonation is easier to recognize than question intonation; 2. the sentence-final tone does not affect statement intonation recognition; 3. question intonation is easier to recognize if the sentence-final tone is falling whereas it is harder to recognize if the sentence-final tone is rising. Implications of the results for the modeling of Chinese intonation are discussed.

Advances in Social Science, Education and Humanities Research, 2022

This paper explores tone2 and tone3 in monosyllabic words in Mandarin, aiming to illustrate the contributing factors that influence the perception of tone2 and tone3 with low-falling contours. Previous studies mainly focus on inflection position and onset or final F0 values. As the pitch value of the inflection point is a vital acoustic parameter for tone3 identification, this study brings in inflection F0 values as one of the focuses. To reveal the crucial features of tone2 and tone3 identification, this study applies a series of synthetic sounds with identification tasks to investigate three factors that influence tone perception in Mandarin, inflection F0 values, final F0 values, and inflection position, both as independent variables and combined variables. Empirical results suggest that an individual variable cannot determine the perception of tone2 and tone3. Only combinations of variables can decisively alter tone identification. The inflection position plays a minor role in identifications of tone2 and tone3, contradicting with findings from previous researches. The status of inflection F0 values seems to be crucial in tone3 perception. The relationship between inflection F0 values and inflection positions is mutually promotion and compensation for tone3 identification. Final F0 values function as compensation for the emphasis on T2 perception caused by inflection F0 values. The perception of tone2 and tone3 does not show a categorical pattern at a stimuli continuum with varying inflection F0 values.

Language and Cognition, 2020

ABSTRACTWe intended to establish if two lexical tone contrasts in Zhumadian Mandarin, one between early and late aligned falls and another between early and late aligned rises, are perceived categorically, while the difference between declarative and interrogative pronunciations of these four tones is perceived gradiently. Presenting stimuli from 7-point acoustic continua between tones and between intonations, we used an identification task and a discrimination task with an experimental group of native listeners and a control group of Indonesian listeners, whose language employs none of the differences within either the falling or the rising pitch contours in its phonology. Only the lexical condition as perceived by the experimental group yielded sigmoid identification functions and a heightened discriminatory sensitivity around the midpoint of continua. The intonational condition in the native group and both conditions in the control group yielded gradient identification functions ...

2020

In connected speech, the acoustic properties of Mandarin tones undergo modifications not observed in isolation. The current study investigated the perceptual distinction between Mandarin tones in connected speech, focusing on Tone 3 and Tone 4, which have been reported to share a similar initial falling contour. The current study also tested whether syllables produced with focus and / or in certain syllable positions affect the tonal perception. In a forced choice perception task, participants heard syllables extracted from three syllable words previously recorded in short dialogues, and were instructed to select one of four characters representing corresponding monosyllabic words differing only in tone. The accuracy results showed that Tone 4 was much more successfully identified than Tone 3. Nonetheless, after using a dprime analysis to control for an observed T4 response bias, we found the same level of perceptibility of T3 and T4. Furthermore, the two tones were better perceived when a tone was produced in a focus context or at the edge of a word, confirming the effect of prosodic structure on tonal perception.

The intonation patterns of Mandarin and Cantonese are very different. While Mandarin uses a raised global F0 contour to signal questions, Cantonese relies on a boundary tone strictly located at the end of a question. Previous individual-language studies and cross-linguistic studies with normal speech stimuli have shown that intonation interpretation was influenced by both universal perceptual factors and language-tone specific factors. In this study, Mandarin and Cantonese listeners participated in a forced identification task to judge the sentence type of the low-pass filtered Mandarin or Cantonese utterances they heard. The results confirm the different intonation patterns between Cantonese and Mandarin. They also show that when lexical-semantic information was absent and when identities of lexical tones were covered up, listeners relied solely on the frequency code to interpret the intonation. Finally, the results also suggest that F0 contour alone was not sufficient for intonation perception. Lexical-semantic context, segmental information and duration-rhythmic profile also contribute to intonation production and perception.

Previous research has not determined whether the inclusion of lexical semantic information facilitates or inhibits the learning of second language (L2) phonetic contrasts. The present study addresses this issue by comparing the acquisition of Mandarin Chinese tones with and without semantic contexts. Two groups of native English listeners with no lexical tone experience participated in a Mandarin tone training program where one ("No meaning") group received training with only phonetic tonal contrasts, while the other ("Meaning") group was additionally provided with semantic information. Results show that although both groups started comparably and improved significantly with training, the "No-meaning" trainees had significantly higher tone identification accuracy rates than the "Meaning" trainees after training. However, the inter-session tests with the training stimuli reveal the opposite pattern, where the Meaning group outperformed the No-meaning group. Together, these results indicate that, at the initial stage of tone learning, non-native listeners learn more efficiently by focusing on phonetic tonal distinctions, whereas remembering the meanings of tone words does not generalize well to tone category identification.