We discuss a representational scheme used in a hybrid naturallanguageprocessing system. The system is called SYMCON (Symbolic and Connectionist), and as its name implies, it consists of subsystems of different natur...
详细信息
We discuss a representational scheme used in a hybrid naturallanguageprocessing system. The system is called SYMCON (Symbolic and Connectionist), and as its name implies, it consists of subsystems of different nature. SYMCON is being developed primarily for the task of word sense disambiguation (WSD). In order to disambiguate different word senses, all the necessary information (e.g., syntactic and semantic information) must firstly be represented in a machine tractable manner. Furthermore, the representations must be effective for the different subsystems in the hybrid SYMCON system to cooperate and work as whole. The representational scheme plays a crucial part in the development of the hybrid system. The system organisation of SYMCON is overviewed. Then important issues of the distributed representational scheme based on microfeatures are discussed in detail.
Fundamentals of fuzzy knowledge base for image understanding are dealt with, It consists of the data-word transformation part (to transform numerical data, derived from image processing, into words) and answer generat...
详细信息
Fundamentals of fuzzy knowledge base for image understanding are dealt with, It consists of the data-word transformation part (to transform numerical data, derived from image processing, into words) and answer generation part (to answer about objects and phenomena in the image world). Ambiguous recognition results are handled by a fuzzy matching and a fuzzy frame knowledge-based system. By applying fuzzy IF-THEN rules, the process of image understanding can be made independent of weather conditions and daylight changing. This system also employs user friendly information retrieval system by applying naturallanguage instructions.< >
Speaker accent is an important issue in the formulation of robust speaker independent recognition systems. knowledge gained from a reliable accent classification approach could improve overall recognition performance....
详细信息
Speaker accent is an important issue in the formulation of robust speaker independent recognition systems. knowledge gained from a reliable accent classification approach could improve overall recognition performance. In this paper, a new algorithm is proposed for foreign accent classification of American English. A series of experimental studies are considered which focus on establishing how speech production is varied to convey accent. The proposed method uses a source generator framework, recently proposed for analysis and recognition of speech under stress [5]. An accent sensitive database is established using speakers of American English with foreign language accents. An initial version of the classification algorithm classified speaker accent from among four different accents with an accuracy of 81.5% in the case of unknown text, and 88.9% assuming known text. Finally, it is shown that as ascent sensitive word count increases, the ability to correctly classify accent also increases, achieving an overall classification rate of 92% among four accent classes.
This paper describes a technique for automatic speaker verification based on prosodic knowledge in Hindi using neural networks. Properties of intonation patterns (changes in F0 as a function of time) and duration were...
详细信息
Automatic speech recognition ( ASR ) systems are made up of a number of different knowledge sources (KSs) which combine to solve the overall problem of speech recognition. An investigation into the benefits of each KS...
详细信息
This paper presents a knowledge-based user interface model system for support of a mining teleoperation. The model makes extensive use of non-verbal sentential structures (NVSS). The NVSS is built with the GURUtm expe...
详细信息
This paper describes effects on utterances caused by knowledge about the hearer. Our study focuses on the difference of expressions in the dialogue of two groups. Two groups of six and seven subjects were asked to obt...
详细信息
This paper presents a parallel approach for utilizing contextual knowledge to improve spoken language understanding. The method emphasizes a hierarchically-structured knowledge base and a memory-based parsing techniqu...
详细信息
This paper presents a parallel approach for utilizing contextual knowledge to improve spoken language understanding. The method emphasizes a hierarchically-structured knowledge base and a memory-based parsing technique. Within this paradigm, several levels of knowledge sources including contextual knowledge arc efficiently combined. An ambiguity resolution scheme utilizing the preceding discourse context and the situational context was implemented on a parallel computer using a marker-passing scheme. The experiments on the parallel computer for an Air Traffic Control (ATC) domain show an 86% sentence recognition accuracy with about 6% improvement compared with the system not utilizing the contextual knowledge.
暂无评论