Expertise in Chinese character recognition is marked by reduced holistic processing, which depends mainly on writing rather than reading experience. Here we show that, while simplified and traditional Chinese readers demonstrated a similar level of HP when processing characters shared between the simplified and traditional scripts, simplified Chinese readers were less holistic than traditional Chinese readers in perceiving simplified characters; this effect depended mainly on their writing rather than reading performance. However, the two groups (...) did not differ in HP of traditional characters, regardless of their difference in reading and writing performances. Our image analysis showed high visual similarity between the two character types, with a larger variance among simplified characters; this may allow simplified Chinese readers to interpolate and generalize their skills to traditional characters. Thus, transfer of perceptual expertise may be constrained by both the similarity in feature and the difference in exemplar variance between the categories. (shrink)

Measuring Chinese character recognition ability is essential in research on character learning among learners of Chinese as a second language. Three methods are typically used to evaluate character recognition competence by investigating the following properties of a given character: pronunciation, meaning, and pronunciation and meaning. However, no study has explored the similar or dissimilar outcomes that these three measurements might yield. The current study examined this issue by testing 162 CSL learners with (...) various L1 backgrounds and Chinese proficiency levels. Participants' performance in character recognition measured using a phonological method, a semantic method, and a PS method was compared, which led to two major findings. In terms of similarity, participants' performance in character recognition and the influence of L1 background and Chinese proficiency level on character recognition was similar across the three methods. As for differences, the semantic method could yield a character recognition test with better quality than the other two methods, and the three methods yielded different best fitting models and showed different predictions for Chinese proficiency across different L1 groups. Theoretical and practical implications of these findings are proposed. (shrink)

In Chinese orthography, the most common character structure consists of a semantic radical on the left and a phonetic radical on the right ; the minority, opposite arrangement also exists. Recent studies showed that SP character processing is more left hemisphere lateralized than PS character processing. Nevertheless, it remains unclear whether this is due to phonetic radical position or character type frequency. Through computational modeling with artificial lexicons, in which we implement a theory of hemispheric (...) asymmetry in perception but do not assume phonological processing being LH lateralized, we show that the difference in character type frequency alone is sufficient to exhibit the effect that the dominant type has a stronger LH lateralization than the minority type. This effect is due to higher visual similarity among characters in the dominant type than the minority type, demonstrating the modulation of visual similarity of words on hemispheric lateralization. (shrink)

This study used priming paradigm with lexical decision task to examine the effects of different levels of semantic relatedness on the identification of Chinese phonetic–semantic compound characters. Unlike previous studies that simply classify Chinese compound characters as semantically transparent or opaque, we categorize the semantic relatedness between semantic radicals and the target characters containing them into five levels: highly related, moderately related, minimally related, unrelated but sharing the semantic radical, and unrelated without sharing the semantic radical. Moreover, three (...) stimulus-onset asynchrony s were used in this study to dissociate the radical- and character-level semantic priming effects. Results revealed a graded priming effect of the semantic radical on character recognition in Chinese readers for all SOAs. More specifically, the facilitative effect of the semantic radical on character processing was most evident for the high condition, followed by the minimal, form-only, and control conditions. This suggests a graded priming effect of the semantic radical on character identification. (shrink)

Long ago the Chinese people developed the habit of thinking in terms of images. They also formed the habit of writing and recognizing scriptforms in terms of images. In fact, these diverse cognitive processes - thinking, writing and decoding in terms of images - have been interacting and reinforcing one another for thousands of years, and, as a result, have played a significant role in shaping Chinese culture and the Chinese mind, and have become a part of (...) the collective unconscious of the people.Although today the Chinese characters are highly abstract and symbolic, the language still preserves its pictographic and ideographic prototypes, the parent script which evolved almost six thousand years ago. In other words, although modern Chinese characters are no longer actual drawings of objects, ideas, and images, they still have a blood relationship to these primitive forms. Their graphemes, configurations, or grapho-semantic radicals still bear direct semantic relation to their referents, so much so that it would be fair to describe the development of Chinese writing as an overlapping process as well as a gradual one. (shrink)

Majority of Chinese characters are pictographic characters with strong associative ability and when a character appears for Chinese readers, they usually associate with the objects, or actions related to the character immediately. Having this background, we propose a system to visualize the simplified Chinese characters, so that developing any skills of either reading or writing Chinese characters is not necessary. Considering the extensive use and application of mobile devices, automatic identification of Chinese characters (...) and display of associative images are made possible in smart devices to facilitate quick overview of a Chinese text. This work is of practical significance considering the research and development of real-time Chinese text recognition, display of associative images and for such users who would like to visualize the text with only images. The proposed Chinese character recognition system and visualization tool is named as MyOcrTool and developed for Android platform. The application recognizes the Chinese characters through OCR engine, and uses the internal voice playback interface to realize the audio functions and display the visual images of Chinese characters in real-time. (shrink)

There is controversy around whether presenting sub-character units such as radicals and strokes are beneficial to L2 Chinese learning. The present study explored the effects of radical markings and stroke order animations on learning Chinese characters. Forty Chinese L2 learners with native alphabetic languages were divided into high-and low-level groups. They were first required to learn Chinese characters under four conditions either: presented radical markings with stroke animations; presented no radical markings with stroke animations; presented (...) radical markings without stroke animations; or presented neither radical markings nor stroke animations. After learning, the participants were given character recognition and character-meaning matching tests. Results showed that the presentation of radical markings increased the participants’ reaction times in the character recognition test and decreased their recognition accuracy. Moreover, presenting stroke order animations also decreased the participants’ accuracy in recognizing characters. Beyond that, presenting radical markings and stroke order animations had no significant influence on character-meaning matching tests. These results indicate that providing radical and stroke information might interfere with character learning instead of facilitating character learning. The results suggest that excessive visual information introduced in the learning process may increase L2 learners’ cognition load. Also, the findings contribute to theoretical arguments about the analytic and holistic processing of Chinese characters and the pedagogical implications for teaching Chinese as a second language. (shrink)

In the area of machine learning, different techniques are used to train machines and perform different tasks like computer vision, data analysis, natural language processing, and speech recognition. Computer vision is one of the main branches where machine learning and deep learning techniques are being applied. Optical character recognition is the ability of a machine to recognize the character of a language. Pashto is one of the most ancient and historical languages of the world, spoken in (...) Afghanistan and Pakistan. OCR application has been developed for various cursive languages like Urdu, Chinese, and Japanese, but very little work is done for the recognition of the Pashto language. When it comes to handwritten character recognition, it becomes more difficult for OCR to recognize the characters as every handwritten character’s shape is influenced by the writer’s hand motion dynamics. The reason for the lack of research in Pashto handwritten character data as compared to other languages is because there is no benchmark dataset available for experimental purposes. This study focuses on the creation of such a dataset, and then for the evaluation purpose, a machine is trained to correctly recognize unseen Pashto handwritten characters. To achieve this objective, a dataset of 43000 images was created. Three Feed Forward Neural Network models with backpropagation algorithm using different Rectified Linear Unit layer configurations were trained and tested with this dataset. The simulation shows that Model 1 achieved accuracy up to 87.6% on unseen data while Model 2 achieved an accuracy of 81.60% and 3% accuracy, respectively. Similarly, loss was the lowest for Model 1 with 0.15 and 3.17 for training and testing, followed by Model 2 with 0.7 and 4.2 for training and testing, while Model 3 was the last with loss values of 6.4 and 3.69. The precision, recall, and f-measure values of Model 1 were better than those of both Model 2 and Model 3. Based on results, Model 1 is found to be the most efficient as compared to the other two models in terms of accuracy to recognize Pashto handwritten characters. (shrink)

The goal of this study is to investigate whether different types of structures and lexical tones of Chinese characters cause different processing fluency. In Experiment 1, participants’ explicit affective assessments of Chinese characters with different structures, frequencies, and lexical tones were analyzed. Results indicated that participants showed explicit preferences and dispreferences to different structures and lexical tones. In Experiment 2, participants’ implicit responses to different structures and lexical tones were investigated using a metaphor experimental paradigm. Results were consistent (...) with the major findings of Experiment 1. In Experiments 3 and 4, participants’ recognition of words of different structures and lexical tones were analyzed. Results revealed that participants had a better memory for Surround structure characters when stimuli were visually presented and for Tone 3 when stimuli were auditorily presented. Finally, the significance and implications of this study are discussed. (shrink)

The effects of psycholinguistic variables on reading development are critical to the evaluation of theories about the reading system. Although we know that the development of reading depends on both individual differences (endogenous) and item-level effects (exogenous), developmental research has focused mostly on average-level performance, ignoring individual differences. We investigated how the development of word recognition in Chinese children in both Chinese and English is affected by (a) item-level, exogenous effects (word frequency, radical consistency, and curricular grade (...) level); (b) subject-level, endogenous individual differences (orthographic awareness and phonological awareness); and (c) their interactive effect. We tested native Chinese (Putonghua)-speaking children (n = 763) in grades 1 to 6 with both Chinese character and English word identification (lexical) decision tasks. Our findings show that (a) there were effects of both word frequency and age of acquisition in both Chinese and English, but these item-level effects generally weakened with increasing age; (b) individual differences in phonological and orthographic awareness each contributed to successful performance; and (c) in Chinese, item-level effects were weaker for more proficient readers. We contend that our findings can be explained by theoretical models that incorporate cumulative learning as the basis for development of item-level effects in the reading system. (shrink)

Presently, the State Grid Corporation of China has accumulated a large amount of maintenance records for power primary equipment. Unfortunately, most of these records are unstructured data which lead to difficultly analyze and utilize them. The emergence of natural language processing technology and deep learning methods provide a solution for unstructured text data. This paper proposes a progressive multitype feature fusion model to recognize Chinese named entity of unstructured maintenance records for power primary equipment. Firstly, the textual characteristics and (...) word separation difficulties of maintenance records are analyzed, then 7 main entity categories of power technical terms from unstructured maintenance records are chosen, and 3452 maintenance records are labeled by these categories, which is so called EPE-MR training dataset. Secondly, the standard test reports, standard maintenance, and fault analysis reports for three types of power primary equipment are employed as corpus to train character embedding in order to obtain certain words representation ability of maintenance records. After that, progressive multilevel radicals feature extraction module is designed to get detailed and fine semantic information in a hierarchical manner. Further, radicals feature representation and character embedding are concatenated and sent to BiLSTM module to extract contextual information in order to improve Chinese entity recognition ability. Moreover, CRF is introduced to handle the dependencies among prediction labels and to output the optimal prediction sequence, which can easily obtain structured data of maintenance records. Finally, comparative experiments on public MSRA dataset, China People’s Daily corpus, and EPE-MR dataset are implemented, respectively, which show the effectiveness of the proposed method. (shrink)

In a digital era that neglects handwriting, the current study is significant because it examines the mechanisms underlying this process. We recruited 9- to 10-year-old Chinese children, who were at an important period of handwriting development, and adult college students, for both behavioral and electroencephalogram experiments. We designed four learning conditions: handwriting Chinese, viewing Chinese, drawing shapes followed by Chinese recognition, and drawing shapes followed by English recognition. Both behavioral and EEG results showed that (...) HC facilitated visual word recognition compared to VC, and behavioral results showed that HC facilitated visual word recognition compared to drawing shapes. HC and VC resulted in a lateralization of the N170 in adults, but not in children. Taken together, the results of the study suggest benefits of handwriting on the neural processing and behavioral performance in response to Chinese characters. The study results argue for maintaining handwriting practices to promote the perception of visual word forms in the digital age. (shrink)

This study adopted a change detection task to investigate whether and how recognition intent affects the construction of orthographic representation in visual word recognition. Chinese readers and nonreaders detected color changes in radical components of Chinese characters. Explicit recognition demand was imposed in Experiment 2 by an additional recognition task. When the recognition was implicit, a bias favoring the radical location informative of character identity was found in Chinese readers , but (...) not nonreaders . With explicit recognition demands, the effect of radical location interacted with radical function and word frequency . An estimate of identification performance under implicit recognition was derived in Experiment 3. These findings reflect the joint influence of recognition intent and orthographic regularity in shaping readers’ orthographic representation. The implication for the role of visual attention in word recognition was also discussed. (shrink)

Unlike in English, the role of phonology in word recognition in Chinese is unclear. In this event-related potential experiment, we investigated the role of phonology in reading both high- and low-frequency two-character compound Chinese words. Participants executed semantic and homophone judgment tasks of the same precede-target pairs. Each pair of either high- or low-frequency words were either unrelated or related semantically or phonologically. The induced P200 component was greater for low- than for high-frequency word-pairs both in (...) semantic and phonological tasks. Homophones in the semantic judgment task and semantically-related words in the phonology task both elicited a smaller N400 than the control condition, word frequency-independently. However, for low-frequency words in the phonological judgment task, it was found that the semantically related pairs released a significantly larger P200 than the control condition. Thus, the semantic activation of both high- and low-frequency words may be no later than phonological activation. (shrink)

The article focuses on the key issues of contemporary Chinese academic publications on the specifcs of Chinese rationality. Among its most common characteristics, there are focus on practice, attention to everyday life of society and individuals, pursuit of centrality and harmony in the moral sphere. Cross-cultural comparisons serve to justify the role of traditional rationality as a balancer keeping Chinese thought away from subjectivism, irrationalism, abstract reasoning and formalism. The researchers seek to identify the impacts of “applied (...) rationality” upon the formation of positive features of Chinese national character. The widespread interpretation of Chinese practical rationality underlines the priority of practice of political governance over the theoretical reasoning about politics; in the moral sphere it leads to rejection of extremes in human behavior and sets the normative status of the middle path and harmony. The inclination of humans to keep their emotions under control and to take a proactive approach to the world was accompanied by recognition of the higher value of empirical experience and intuition compared to formal and abstract ideas. The focus on the secular life of people deprived the Chinese rationality of the “otherworldly” religious dimension. The participants of the discussion note that the Chinese culture has gained stability and “elasticity,” but it lacks Western formal rationality, which is a prerequisite for scientifc knowledge. Chinese researchers pay much attention to the impact of traditional rationality on the formation of Chinese philosophy in the 20th century and to the choice of socio-political path of national development. The emphasis on the differences between the cultures ofChinaand the West is combined with searches for similarities between Chinese rationality and Marxism. That steers the discussion beyond the sphere of comparative studies and makes its content relevant to the understanding of broader set of contemporary problems. (shrink)

Studies have suggested that visually presented words are obligatorily decomposed into constituents that could be mapped to language representations. The present study aims to elucidate how orthographic processing of one constituent affects the other and vice versa during a word recognition task. Chinese orthographic system has characters representing syllables and meanings instead of suffixation roles, and the majority of Chinese characters are phonograms that can be further decomposed into phonetic radical and semantic radical. We propose that semantic (...) radical combinability indexed by semantic radicals and the effect of phonological consistency indexed by phonetic radicals would interact with each other during the reading of Chinese phonograms. Twenty-six right-handed native Chinese speakers were recruited to the study. Participants were presented with phonograms divided into four conditions following their semantic radical combinability and phonological consistency. EEG signals were recorded throughout the covert naming task. Our results show that there is an interaction effect between phonological consistency and semantic radical combinability on the right hemisphere N170 activity while reading phonograms. Semantic radical combinability influenced the right hemisphere N170 during the process of low-consistency character reading but not high-consistency character reading. On the other hand, the left hemisphere N170 revealed a more significant activity during reading high-consistency characters and was not affected by radical combinability. In addition, while low-consistency characters revealed a larger P200 than high-consistency characters, the semantic radical combinability effect on P200 was only significant when participants were reading high-consistency characters but not low-consistency characters. These results provide new information about how ERPs are involved in word recognition within the context of interaction among orthographic and phonological dimensions. (shrink)

Information on age-related differences in affective meanings of words is widely used by researchers to study emotions, word recognition, attention, memory, and text-based sentiment analysis. To date, no Chinese affective norms for older adults are available although Chinese as a spoken language has the largest population in the world. This article presents the first large-scale age-related affective norms for 2,061 four-character Chinese words (AANC). Each word in this database has rating values in the four dimensions, (...) namely, valence, arousal, dominance, and familiarity. We found that older adults tended to perceive positive words as more arousing and less controllable and evaluate negative words as less arousing and more controllable than younger adults did. This indicates that the positivity effect is reliable for older adults who show a processing bias toward positive vs. negative words. Our AANC database supplies valuable information for researchers to study how emotional characteristics of words influence the cognitive processes and how this influence evolves with age. This age-related difference study on affective norms not only provides a tool for cognitive science, gerontology, and psychology in experimental studies but also serves as a valuable resource for affective analysis in various natural language processing applications. (shrink)

Leibniz was interested in China throughout his life, and he admired its culture. Originally, his interests revolved around Chinese characters, but widened when meeting the Jesuit China missionary P. Grimaldi in Rome 1689. From that time on, Leibniz pursued the project of a knowledge exchange between both sides of the world. He was convinced that Europe and China were on the same cultural level, while diverging over advances in distinct fields. In his view, Europe was more advanced in theoretical (...) areas, such as logic, mathematics, geometry, physics, mathematical astronomy, and theory of natural religion, while China was superior in empirical disciplines, such as in observational astronomy, medicine, historiography, and moral philosophy, which he understood as natural (i.e. rational) theology. This project of knowledge exchange presupposed the recognition of Chinese culture. In this sense, Leibniz strongly supported the Jesuit position, according to which the rites of ancestral worship and reverence of Confucius are riti politici rather than literal expressions of religious faith. Leibniz’s main objection to the Pope’s negative assessment of Chinese rites was that one cannot settle the so-called Rites Controversy as long as one fails to interpret adequately Chinese culture and Confucian thought in particular. (shrink)

Synaesthesia is a neurological condition in which a sensory or cognitive stimulus consistently co-activates another sensory/cognitive quality in addition to its usual qualities. For example, synaesthetes might see colours when they read words. Coloured language is one of the most common and most studied types of synaesthesia. The processes that govern the associations of colours and language have been linked to the mechanisms underlying the processing of language more generally. This chapter reviews evidence from current psycholinguistic synaesthesia research in (...) class='Hi'>Chinese, focusing in particular on how synaesthetic colouring works in Chinese characters, the basic writing units in Chinese. I consider this in relation to the psycholinguistic processes of character recognition. This chapter reviews evidence for the following facts: that synaesthetic colouring of Chinese characters is a genuine phenomenon in the Chinese population and may affect as many as 1 in 100 in the Chinese population, with a non-significant difference across the sexes that the colouring of characters is rule-based and influenced by a number of linguistic factors; and that synaesthetic colouring can transfer across Chinese and English in bilinguals. These findings are discussed in relation to native versus non-native Chinese synaesthetes, and to the Chinese versus English systems. (shrink)

Writing, or calligraphy, in China is strongly influenced by ancient techniques of making art. Chinese characters have evolved from the patterns of bronze drawings, and China’s earliest hieroglyphs usually retain the traces of their origin in paintings. These paintings usually recorded daily life, and the related Chinese characters have evolved from these with general, simplified and abstract features. The composition that makes Chinese characters is a manifestation of ancient Chinese philosophy, of which Confucianism, Taoism and Buddhism (...) are the three key strands. They have affected Chinese culture and aesthetics in all dynasties. Chinese philosophy emphasizes the integration of individuals in their surrounding environment. The art of ancient China consequently combines skills that relate to the artificial, or constructed, and to natural patterns. Chinese characters make a perfect example of this. The article/presentation will outline the research that I have so far undertaken on the relation between Chinese calligraphy and architectural space, between writing and constructing environments. It will also present some experimental design research that aims at developing from the above-mentioned basic research new architectural typologies that are contemporary, resonant and sensitive to the Chinese context. (shrink)

This article makes a preliminary attempt to account for the stroke systems of Chinese characters in simplified regular script. The framework utilized is the three meta-functions in Systemic Functional Linguistics. The description observes the cases from the perspectives of the experiential, appraisal, and thematic semiosis of strokes and their constitutional segments to figure out the relevant systems: the line system and the point system. This process witnesses comparisons to seek, in brief though, the traces and origins of stroke development (...) along evolution, and hence the straightening of lines and squaring of character formation. This is the first step towards a rank model of Chinese writing, the whole project of which will highlight a way to study other writing systems in the semiotic respect. (shrink)

In light of the dramatic growth of Chinese learners worldwide and a need for cross-linguistic research on Chinese literacy development, this study drew upon theories of visual complexity effect (Su & Samuels, 2010) and dual-coding processing (Sadoski & Paivio, 2013) and investigated a) the effects of character properties (i.e., visual complexity and radical presence) on character acquisition and b) the relationship between individual learner differences in radical awareness and character acquisition. Participants included adolescent English-speaking beginning (...) learners of Chinese in the U.S. Following Kuo et al. (2014), a novel character acquisition task was used to investigate the process of acquiring the meaning of new characters. Results showed that a) characters with radicals and with less visual complexity were easier to acquire than characters without radicals and with greater visual complexity; and b) individual differences in radical awareness were associated with the acquisition of all types of characters, but the association was more pronounced with the acquisition of characters with radicals. Theoretical and practical implications of the findings were discussed. (shrink)

It has not been clarified if attention influences perception of targets in visual masking. Three forms of common masks were thus chosen in the present study and presented with character targets in three temporal sequences. In order to pinpoint the level of processing where masking arises, character targets were varied in depth of processing from random arrangements of strokes up to real Chinese characters. The attentional influence was examined under perceptual discrimination and lexical decision tasks, respectively. The (...) results revealed significant interactions among four factors. Identification of character targets in each form of mask sequence varied with task demand, with greater suppression in the perceptual discrimination task. These findings suggested that attentional demand can bias processing in favor of task-related information in visual masking. Variations in masking effects may be contributed by both attentional demand and spatio-temporal interaction. (shrink)

Lexical ambiguity is pervasive in language, and the nature of the representations of an ambiguous word's multiple meanings is yet to be fully understood. With a special focus on Chinese characters, the present study first established that native speaker's perception about a character's number of meanings was heavily influenced by the availability of its distinct word formations, while whether these meanings would be perceived to be closely related was driven by further conceptual analysis. These notions were operationalized as (...) two computed metrics, which assessed the degree of dispersion across individual word formations and the degree of propinquity across clusters of word formations, respectively, in a distributional semantic space. The observed correlations between the computed and the perceived metrics indicated that the utility of word formations to tap into meaning representations of Chinese characters was indeed cognitively plausible. The results have demonstrated the extent to which distributional semantics could inform about meaning representations of Chinese characters, which has theoretical implications for the representation of ambiguous words more generally. (shrink)

Chinese characters are central to understanding how learners learn to read a logographic script. However, researchers know little about the role of character writing in reading Chinese as a second language. Unlike an alphabetic script, a Chinese character symbol transmits semantic information and is a cultural icon bridging embodied experience and text meaning. As a unique embodied practice, writing by hand contributes to cognitive processing in Chinese reading. Therefore, it is essential to clarify how (...) Chinese character writing, language distance, and cultural background influence CSL reading proficiency. Based on extant research on L2 reading acquisition and strength of key theoretical perspectives of embodied cognition theory, this study tested a regression model for CSL reading involving individual-level factors and group-level predictors. This study collected big data in a sample of 74,362 CSL learners with 67 diverse L1s. Results of hierarchical linear modeling showed a significant effect of CCWP and significant language distance × CCWP interaction effect on reading proficiency; however, cultural background × CCWP interaction effect was not significant. These results conform to the ECT and indicate that bodily activity, past language usage, and cultural background aided reading. CCWP may benefit from withstanding the negative transfer from L1s. Furthermore, CCWP and cultural background are not synergistic predictors of reading. This study may open novel avenues for explorations of CSL reading development. (shrink)

Handwritten text recognition is considered as the most challenging task for the research community due to slight change in different characters’ shape in handwritten documents. The unavailability of a standard dataset makes it vaguer in nature for the researchers to work on. To address these problems, this paper presents an optical character recognition system for the recognition of offline Pashto characters. The problem of the unavailability of a standard handwritten Pashto characters database is addressed by developing (...) a medium-sized database of offline Pashto characters. This database consists of 11352 character images. Enriched feature extraction techniques of histogram of oriented gradients and zoning-based density features are used for feature extraction of carved Pashto characters. K-nearest neighbors is considered as a classification tool for the proposed algorithm based on the proposed feature sets. A resultant accuracy of 80.34% is calculated for the histogram of oriented gradients, while for zoning-based density features, 76.42% is achieved using 10-fold cross validation. (shrink)

Speech recognition technology has played an indispensable role in realizing human-computer intelligent interaction. However, most of the current Chinese speech recognition systems are provided online or offline models with low accuracy and poor performance. To improve the performance of offline Chinese speech recognition, we propose a hybrid acoustic model of deep convolutional neural network, long short-term memory, and deep neural network. This model utilizes DCNN to reduce frequency variation and adds a batch normalization layer after (...) its convolutional layer to ensure the stability of data distribution, and then use LSTM to effectively solve the gradient vanishing problem. Finally, the fully connected structure of DNN is utilized to efficiently map the input features into a separable space, which is helpful for data classification. Therefore, leveraging the strengths of DCNN, LSTM, and DNN by combining them into a unified architecture can effectively improve speech recognition performance. Our model was tested on the open Chinese speech database THCHS-30 released by the Center for Speech and Language Technology of Tsinghua University, and it was concluded that the DLD model with 3 layers of LSTM and 3 layers of DNN had the best performance, reaching 13.49% of words error rate. (shrink)

Grapheme-colour synaesthetes experience unusual colour percepts when they encounter letters and/or digits. Studies of English-speaking grapheme-colour synaesthetes have shown that synaesthetic colours are sometimes triggered by rule-based linguistic mechanisms . In contrast, little is known about synaesthesia in logographic languages such as Chinese. The current study shows the mechanisms by which synaesthetic speakers of Chinese colour their language. One hypothesis is that Chinese characters might be coloured by their constituent morphological units, known as radicals, and we tested (...) this by eliciting synaesthetic colours for characters while manipulating features of the radicals within them. We found that both the function and position of radicals influence the nature of the synaesthetic colour generated. Our data show that in Chinese, as in English, synaesthetic colours are influenced by systematic rules, rather than by random associations, and that these rules are based on existing psycholinguistic mechanisms of language processing. (shrink)

Studies of linguistic synaesthesias in English have shown a range of fine-grained language mechanisms governing the associations between colours on the one hand, and graphemes, phonemes and words on the other. However, virtually nothing is known about how synaesthetic colouring might operate in non-alphabetic systems. The current study shows how synaesthetic speakers of Mandarin Chinese come to colour the logographic units of their language. Both native and non-native Chinese speakers experienced synaesthetic colours for characters, and for words spelled (...) in the Chinese spelling systems of Pinyin and Bopomo. We assessed the influences of lexical tone and Pinyin/Bopomo spelling and showed that synaesthetic colours are assigned to Chinese words in a non-random fashion. Our data show that Chinese-speaking synaesthetes with very different native languages can exhibit both differences and similarities in the ways in which they come to colour their Chinese words. (shrink)