Information geometry theory of high-dimension space and application for speaker independent continuous digit speech recognition

In the light of descriptive geometry and notions in set theory, this paper re-defines the basic elements in space such as curve and surface and so on, presents some fundamental notions with respect to the point cover based on the High-dimension space (HDS) point covering theory, finally takes points from mapping part of speech signals to HDS, so as to analyze distribution information of these speech points in HDS, and various geometric covering objects for speech points and their relationship. Besides, this paper also proposes a new algorithm for speaker independent continuous digit speech recognition based on the HDS point dynamic searching theory without end-points detection and segmentation. First from the different digit syllables in real continuous digit speech, we establish the covering area in feature space for continuous speech. During recognition, we make use of the point covering dynamic searching theory in HDS to do recognition, and then get the satisfying recognized results. At last, compared to HMM (Hidden Markov models)-based method, from the development trend of the comparing results, as sample amount increasing, the difference of recognition rate between two methods will decrease slowly, while sample amount approaching to be very large, two recognition rates all close to 100% little by little. As seen from the results, the recognition rate of HDS point covering method is higher than that of in HMM (Hidden Markov models) based method, because, the point covering describes the morphological distribution for speech in HDS, whereas HMM-based method is only a probability distribution, whose accuracy is certainly inferior to point covering.

Study on Closed-set Speaker Identification Based on Biomimetic Pattern Recognition

In this paper, a new classifier of speaker identification has been proposed, which is based on Biomimetic pattern recognition (BPR). Distinguished from traditional speaker recognition methods, such as DWT, HMM, GMM, SVM and so on, the proposed classifier is constructed by some finite sub-space which is reasonable covering of the points in high dimensional space according to distributing characteristic of speech feature points. It has been used in the system of speaker identification. Experiment results show that better effect could be obtained especially with lesser samples. Furthermore, the proposed classifier employs a much simpler modeling structure as compared to the GMM. In addition, the basic idea "cognition" of Biomimetic pattern recognition (BPR) results in no requirement of retraining the old system for enrolling new speakers.

基于组块及记忆的词性自动标注

基于组块及记忆的模型(BMM)采用与传统方法明显不同的标注思路,以汉语中的整句为处理单元,从组块出发,立足于单个词汇,分析更为丰富的上下文语境知识,并借助知网词典记忆词性集合,同时采用渐增式的机械学习方式获取参数值。对于棘手的稀疏数据问题只简单地设置平伏常数加以平滑,最后利用少量人工规则修正标注结果。实验表明,该模型的封闭式测试准确率将近99%,开放式测试准确率为95%以上。

语音用户界面研究进展

语音是人们日常生活中高效、自然的交流方式之一。但是直到目前为止,语音交互方式在计算机技术上的应用还是比较少的。近年来,随着Ubiquitous Computing和便携式计算机的出现,再次对语音用户界面的应用提出了迫切的需求。而且语音识别、合成技术的发展也为语音交互界面的实现提供了技术基础。本文综合参考了国内外语音界面的一些应用系统实例以及语音这种独特的交流媒体的优点和局限性.总结了语音用户界面的适用环境和设计指导原则,并提出了对语音界面的发展展望。

基于统计的汉语词性标注方法的分析与改进

从词性概率矩阵与词汇概率矩阵的结构和数值变化等方面 ,对目前常用的基于统计的汉语词性标注方法中训练语料规模与标注正确率之间所存在的非线性关系作了分析 .为了充分利用训练语料库 ,提高标注正确率 ,从利用词语相关的语法属性和加强对未知词的处理两个方面加以改进 ,提高了标注性能 .封闭测试和开放测试的正确率分别达到 96.5%和 96% .

语前聋小儿人工耳蜗植入术后的言语康复

An Approach to the Rehabilitation of Prelingually Deaf Children After Cochlear Implantation Zheng Xiujin(Medical Psychology) Directed by Professor Yin WenGang Abstract Objective: To sum up the acquirement rule of speech and language capability which is for the prelingually deaf children after cochlear implantation by listening and language rehabilitation training and to investigate the factors that affect rehabilitation speed. Method: Sixty-four children received a cochlear implant at the age of 2 to 5 years from 2001 to 2005. They begin to be trained under group pattern after switch on 1 month. The whole training program lasted more than 7 months; after that, according to the teacher’s plan the training program was to be continued at home. Result: The period is 108±7.7 days that they can pronounce correctly 50 percent of all of simple-finals and compound-finals, the period is 115.0±7.8 days that they begin auditory repeating, the period is 135.3±10.9 days that they can speech the first specific word independently and the period is 200.3±13.9 days that they can speak 70 words and come into tri-gamut-word and two-word sentence period. The patient that is the group at the age of 2-3 years can take part in normal kindergarten after switch on about 10 months. There are no significant differences in various grades of speech-language development with different age groups and so do with different sex groups. There are significant differences in various grade of speech-language development with various IQ group (P<0.01) and so do with using and not using hearing aids before implantation. Conclusion: From the research we find that the speech and language development sequence is the same level between the prelingually deaf children of 2 to 5 years who received cochlear implant after speech training and normal children and which are stages of uncomplicated sound production, continuous syllabic (babbling), speech sprout, single-word utterances and two-word utterances in proper order. The time is short significantly and the reason is that cognition capability is enhanced along with the increase of age. The intelligence is main factor that affect rehabilitation speed and the speed in the group of high IQ is faster than common IQ. It is not because of the dominance cognition of the senior group that makes the increasing of the rehabilitation, it even makes slowly. The reason of which is that the senior group are exposed the language environment too late to achieve speech and language development. So we should perform an operation and training early. The effectiveness of rehabilitation after cochlear implantation is improved by using hearing aids before implantation. The reason is auditory stimulate can be benefit of to deaf children. The rehabilitation speeds in the children at the age of 2 to 5 years have nothing to do with sex. Key words: cochlear implant; speech therapy; paediatric rehabilitation

汉语文本语篇语句焦点确定及焦点与重音关系研究

This study investigated the method of the focus identification in Chinese text discourse and the relationship between accent and focus, large corpus analysis and decision tree were used in the research. The main results are: 1. Based on the concept of the Focus and understanding of the discourse, Foci identification is consistent and steady; 2. Special Focus markers and specific Focus constructions have greater influence than special constituent order on identifying Focus in Chinese discourse; while information states also have great influence on focus identifying; part of speech，information state, the relative position in the sentence, focus-sensitive operator, specific Focus constructions, contrast relations, relations between the sentences are important factors to focus identifying; 3. Using multi-dimensional tagging and knowledge discovery, it is a feasible way to construct and employ decision trees by computing tagging results to identify Focus; 4. Focus predicting also depends on literal types and styles of the discourse, several types of decision trees should be constructed for different literal types; 5. In the monologue discourse, the most prominent accent is located on the Focus word or in the scope of the Focus; there are some kinds of rules on accent assignment in broad Focus; it is necessary to analyze and classify focus structure for the research of relations between accent and Focus.

语音意识缺陷儿童的言语知觉技能

Considerable studies find that developmental dyslexia is associated with deficits in phonological processing skills, especially phonological awareness. In order to explore the nature of phonological awareness deficits in dyslexia, researchers have begun to investigate the role of speech perception. The findings about speech perception abilities in dyslexics are inconsistent. The heterogeneity of dyslexia may be responsible for the inconsistency of findings. Considering the general suggestion that phonological awareness deficits in dyslexia are attributed to categorical perception deficits, it is more direct to examine whether children with phonological awareness difficulties or phonological dyslexia show speech categorization deficits consistently. The present study would investigate whether Chinese children with phonological awareness deficits or phonological dyslexia showed abnormal speech perception. The whole study consisted of two parts. Part I screened children with phonological-awareness deficits from Year 3 kindergartens and examined their abilities of perceiving native category continuum, nonnative category contrasts and non-speech sound series. Part II selected phonological dyslexics from an elementary school as participants, and further explored the relation between phonological deficits and speech perception. The first two experiments of Part II examined separately the abilities to label stimuli in native category continuum and brief stops in different contexts, the last experiment investigated the adaptation effects of different participant groups. The main conclusions are as follows: 1) Children with phonological dyslexia showed categorical perception deficits: they had lower consistency than controls when perceiving stimuli within phonetic categories, especially for the stimuli which were not natural sounds. 2) Children with phonological dyslexia exhibited a general difficulty of perceiving brief segments of stops from different contexts. 3) Children with phonological dyslexia did not show adaptation to repeatedly presented stimuli. Based on the present conclusions and the findings of previous studies, we suggested that the representations of sound stimuli in phonological dyslexics’ brains are different from those in normal children’s; the representations of sound stimuli in dyslexics’ cortical neural networks are more diffuse and inconsistent.

手持通信设备的多通道文本输入研究

Recently，Handheld Communication Devices is developing very fast, extending in users and spreading in application fields, and has an promising future. This study investigated the acceptance of the multimodal text entry method and the behavioral characteristics when using it. Based on the general information process model of a bimodal system and the human factor studies about the multimodal map system, the present study mainly focused on the hand-speech bimodal text entry method. For acceptance, the study investigated the subjective perception of the accuracy of speech recognition by Wizard of Oz (WOz) experiment and a questionnaire. Results showed that there was a linear relationship between the speech recognition accuracy and the subjective accuracy. Furthermore, as the familiarity increasing, the difference between the acceptable accuracy and the subjective accuracy gradually decreased. In addition, the similarity of meaning between the outcome of speech recognition and the correct sentences was an important referential criterion. The second study investigated three aspects of the bimodal text entry method, including input, error recovery and modal shifts. The first experiment aimed to find the behavioral characteristics of user when doing error recovery task. Results indicated that participants preferred to correct the error by handwriting, which had no relationship with the input modality. The second experiment aimed to discover the behavioral characteristics of users when doing text entry in various types of text. Results showed that users preferred to speech input in both words and sentences conditions, which was highly consistent among individuals, while no significant difference was found between handwriting and speech input in the character condition. Participants used more direct strategy than jumping strategy to deal with mixed text, especially for the Chinese-English mixed type. The third experiment examined the cognitive load in the different modal shifts, results suggesting that there were significant differences between different shifts. Moreover, relevant little time was needed in the Shift from speech input to hand input. Based on the main findings, implications were discussed as follows: Firstly, when evaluating a speech recognition system, attention should be paid to the fact that the speech recognition accuracy was not equal to the subjective accuracy. Secondly, in order to make a speech input system more acceptable, a good method is to train and supply the feedback for the accuracy in training, which improving the familiarity and sensitivity to the system. Thirdly, both the universal and individual behavioral patterns were taken into consideration to improve the error recovery method. Fourthly, easing the study and the use of speech input, the operations of speech input should be simpler. Fifthly, more convenient text input method for non-Chinese text entry should be provided. Finally, the shifting time between hand input and speech input provides an important parameter for the design of automatic-evoked speech recognition system.

语句句法成分边界和韵律特征知觉

The research investigates the acoustic-phonetic correlates of various levels of syntactic boundaries and the perception of prosody in Mandarin Chinese, more specifically, the way speakers express the syntatic relations between sentence compounents and teh perceptual representations of prosody. The relation between phonology and syntax in Chinese language is studied by comparing the perceptual representations and syntactic structures of sentences. The results may have theoretical and practical implications for research in fields of speech perception, linguistics and psycholinguistics, and for the development of speech engineering in China.