The previously proposed syllable-synchronous network search (SSNS) algorithm plays a very important role in the word decoding of the continuous Chinese speech recognition and achieves satisfying performance. Several r...The previously proposed syllable-synchronous network search (SSNS) algorithm plays a very important role in the word decoding of the continuous Chinese speech recognition and achieves satisfying performance. Several related key factors that may affect the overall word decoding effect are carefully studied in this paper, including the perfecting of the vocabulary, the big-discount Turing re-estimating of the N-Gram probabilities, and the managing of the searching path buffers. Based on these discussions, corresponding approaches to improving the SSNS algorithm are proposed. Compared with the previous version of SSNS algorithm, the new version decreases the Chinese character error rate (CCER) in the word decoding by 42.1% across a database consisting of a large number of testing sentences (syllable strings).展开更多
A fundamental difference among modern languages in the world are made by word-syllable structures(WSS), not by distinctive phonemes. Language diversity is supposed to be an evolutionary result of the WSSs, which is de...A fundamental difference among modern languages in the world are made by word-syllable structures(WSS), not by distinctive phonemes. Language diversity is supposed to be an evolutionary result of the WSSs, which is decided by types of syllable constitution and the length of word by syllables. Here we use Swadesh lists of 179 modern languages to analyze their geographic distribution of WSS diversity index and try to discover their developing positions and depths in the evolutionary processes. We also set an ideal WSS offset model for languages, calculate the offset distance and offset direction of each language, and then divide languages into three groups according to the data result, each of which represents an evolutionary type. Our conclusion is that the WSS diversity and the WSS offset model represent the evolutionary trend of diversity and the evolutionary process of human languages in the world. In addition, every language nowadays keeps the most primary WSS features to some extent. Therefore, the WSS may be regarded as genetic factors of human languages.展开更多
文摘The previously proposed syllable-synchronous network search (SSNS) algorithm plays a very important role in the word decoding of the continuous Chinese speech recognition and achieves satisfying performance. Several related key factors that may affect the overall word decoding effect are carefully studied in this paper, including the perfecting of the vocabulary, the big-discount Turing re-estimating of the N-Gram probabilities, and the managing of the searching path buffers. Based on these discussions, corresponding approaches to improving the SSNS algorithm are proposed. Compared with the previous version of SSNS algorithm, the new version decreases the Chinese character error rate (CCER) in the word decoding by 42.1% across a database consisting of a large number of testing sentences (syllable strings).
基金supported by the National Natural Science Fundation of China (31271337)the National Social Science Foundation of China (12&ZD174)
文摘A fundamental difference among modern languages in the world are made by word-syllable structures(WSS), not by distinctive phonemes. Language diversity is supposed to be an evolutionary result of the WSSs, which is decided by types of syllable constitution and the length of word by syllables. Here we use Swadesh lists of 179 modern languages to analyze their geographic distribution of WSS diversity index and try to discover their developing positions and depths in the evolutionary processes. We also set an ideal WSS offset model for languages, calculate the offset distance and offset direction of each language, and then divide languages into three groups according to the data result, each of which represents an evolutionary type. Our conclusion is that the WSS diversity and the WSS offset model represent the evolutionary trend of diversity and the evolutionary process of human languages in the world. In addition, every language nowadays keeps the most primary WSS features to some extent. Therefore, the WSS may be regarded as genetic factors of human languages.
