Method and system for expanding corpus of spoken dialogue system

Abstract

The invention discloses a method for extending spoken language dialogue system corpora. The method includes the steps that secondary semantic analysis is conducted on a sentence which cannot be analyzed to obtain a candidate analysis result; if a user selects a candidate from the candidate analysis result, a mapping rule is formed between the sentence and the candidate selected by the user, the generated rule is added into a user exclusive preprocessing rule library, and the candidate sentence selected by the user and corresponding semantic information are stored in an exclusive auxiliary corpus corresponding to the user; when the user sentence is analyzed, semantic analysis is conducted by means of the user exclusive preprocessing rule library and a rule-assisted main corpus generated on the basis of the auxiliary corpus. The invention further discloses a corresponding system. By means of the method and system, robustness of the corpora can be improved, the cost for maintaining the corpora is reduced, a spoken language system correction function is provided, and the usability of the spoken language system is enhanced.

Description

technical field [0001] This application relates to the expansion technology of the corpus in the spoken language dialogue system, in particular to the method and system for expanding the corpus of the spoken language dialogue system. Background technique [0002] A spoken dialogue system is a computer system that can communicate with people through voice. Spoken language dialogue system is a kind of dialogue system. Compared with general text dialogue system, it mainly has speech recognition and speech synthesis modules. [0003] The dialog system mainly consists of input recognizer / decoder, natural language understanding unit, dialog manager, task managers, natural It consists of a Natural LanguageGenerator unit and an output renderer. [0004] The core of the spoken dialogue system is the natural language understanding unit, which often contains a huge text corpus and consists of three main modules: proper noun recognition, part-of-speech tagging and semantic parser (Sem...

AI Technical Summary

Problems solved by technology

[0010] In practical applications, the biggest problem with the natural language understanding unit is that due to the complexity of the language, the system cannot cover all grammars, and user sentences that are misrecognized or semantically incomprehensible still often appear

In particular, due to the differences in dialects, regional language habits, and user habits, it brings many difficulties to the establishment of the corpus of the spoken dialogue system. Often a very large corpus cannot correctly parse all the words spoken by users.

Therefore, even if the speech recognition system can recognize the sentence spoken by the user, after entering the natural language understanding unit, the semantic parser may not be able to find the corresponding rules to parse the sentence, resulting in a low overall recognition rate

Therefore, developers need to spend a lot of time updating and maintaining the corpus in order to ensure a high recognition rate

Images