Abstract: A system and method for expanding new sample seeds to automatically expand corpora, in which sample seeds are used to collect corpus is provided. The new sample seeds are generated based on the already existed sample seeds and collected corpora; The corpus expansion strategy is determined based on all the sample seeds having been used and new sample seeds: The new sample seeds are refined based on the corpus expansion strategy, and the refined new sample seeds are used to further collect corpus. The above steps are repeatedly executed until predefined condition is satisfied. According to the invention, corpus may be automatically expanded from the web or other resources with low cost and in convenient way to improve the coverage of corpora.

Abstract: A system and method for expanding new sample seeds to automatically expand corpora, in which sample seeds are used to collect corpus is provided. The new sample seeds are generated based on the already existed sample seeds and collected corpora; The corpus expansion strategy is determined based on all the sample seeds having been used and new sample seeds: The new sample seeds are refined based on the corpus expansion strategy, and the refined new sample seeds are used to further collect corpus. The above steps are repeatedly executed until predefined condition is satisfied. According to the invention, corpus may be automatically expanded from the web or other resources with low cost and in convenient way to improve the coverage of corpora.

Abstract: A system and method for expanding new sample seeds to automatically expand corpora, in which sample seeds are used to collect corpus is provided. The new sample seeds are generated based on the already existed sample seeds and collected corpora; The corpus expansion strategy is determined based on all the sample seeds having been used and new sample seeds: The new sample seeds are refmed based on the corpus expansion strategy, and the refmed new sample seeds are used to further collect corpus. The above steps are repeatedly executed until predefined condition is satisfied. According to the invention, corpus may be automatically expanded from the web or other resources with low cost and in convenient way to improve the coverage of corpora.

Abstract: The invention discloses a method for correcting one or more error characters in results of speech recognition, the method comprising the steps of: marking the one or more error characters in the speech recognition results; inputting one or more correct characters corresponding to the one or more marked error characters by input based on character-shape; recognizing the input based on character-shape; displaying one or more candidate characters; selecting one or more desired characters from the one or more candidate characters in accordance with the user; and replacing the one or more error characters with the one or more selected characters; the method characterized by further comprising the step of filtering the one or more candidate characters in accordance with acoustic information associated with the one or more error characters.

Abstract: A method and an apparatus for automatic recognition of tone languages, employing the steps of converting the words of speech into an electrical signal, generating spectral features from the electrical signal, extracting pitch values from the electrical signal, combining said spectral features and the pitch values into acoustic feature vectors, comparing the acoustic feature vectors with prototypes of phonemes in an acoustic prototype database including prototypes of toned vowels to produce labels, and matching the labels to text using a decoder comprising a phonetic vocabulary and a language model database.

Abstract: A Chinese speech recognition (SR) method and system for single or un-correlated Chinese character(s). The method uses various types of Character Description Language (CDL) to describe the single or un-correlated Chinese character(s) to be inputted. The SR system uses CDL grammar directed speech recognizer to accept CDLs, which are inputted by voice. On the basis of analysis of CDL parser, the character generator gives a corresponding character. Therefore, recognition of single or un-correlated Chinese character(s) out of context can be made reliably.