Speech recognition dictionary creation device and speech recognition device

A sound recognition and dictionary technology, applied in speech recognition, speech analysis, instruments, etc., can solve the problems that the approximation degree cannot reach the appropriate value, the number of representations cannot be properly compressed, and the influence of Japanese naturalness is not considered.

Inactive Publication Date: 2005-12-21
PANASONIC INTELLECTUAL PROPERTY CORP OF AMERICA
View PDF1 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, for example, in the case where the voice of "きんどら" is uttered in consideration of abbreviating "金千ドドラマ", the main reason for deciding to generate the expression of changing the word is not considering the words used in combination other than the words used in combination. The number of phonemes extracted from words and the influence of the naturalness of Japanese as a connection of various phonemes
Therefore, there is a problem that the approximation of the performance of changing the statement is not up to the appropriate value
The above-mentioned conventional voice recognition dictionary creation method does not control the expression generation of the changed expression in consideration of the use history of the changed expression expression, so there is a problem that the generated and registered in the recognition dictionary cannot be appropriately compressed. The number of representations that change the statement

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Speech recognition dictionary creation device and speech recognition device
  • Speech recognition dictionary creation device and speech recognition device
  • Speech recognition dictionary creation device and speech recognition device

Examples

Experimental program
Comparison scheme
Effect test

no. 1 Embodiment approach

[0047] figure 1 It is a functional block diagram showing the configuration of the speech recognition dictionary creation device 10 in the first embodiment of the present invention. This voice recognition dictionary compiling device 10 is a device that generates its abbreviations from the recognition target words and registers them as a dictionary, and includes: a recognition target word analysis part 1 and an abbreviation generation part 7 realized as a program or a logic circuit, and a hard disk An analysis word dictionary storage unit 4 , an analysis rule storage unit 5 , an abbreviation generation rule storage unit 6 , and a vocabulary storage unit 8 realized by a storage device such as a non-volatile memory.

[0048] The analysis word dictionary storage unit 4 stores in advance a dictionary related to definitions of unit words (morphological elements) and their phonological series (phonological information) for classifying the recognition target language into structural wo...

no. 2 Embodiment approach

[0071] The second embodiment relates to an example of a voice recognition device incorporating the voice recognition dictionary creation device 10 of the first embodiment and using the voice recognition dictionary 8 a created by the voice recognition dictionary creation device 10 . Embodiments of the present invention relate to such a voice recognition device, which has a dictionary modification function that automatically extracts a word to be recognized from character string information and stores it in a dictionary for voice recognition. The generation of abbreviations is controlled based on information on the history of the abbreviations, and therefore, there is a function of suppressing registration of abbreviations that are less likely to be used in the speech recognition dictionary. Moreover, the so-called character string information refers to information including words (recognition object words) that are recognized by the voice recognition device. For example, the pr...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The high-efficiency compilation of the present invention can also efficiently recognize the voice recognition dictionary compiling device (10) for the expression of changing the expression of the omitted word, and has: a word division part (2), which is composed of one or more than one The recognition target language that the word that forms is divided into structural words; The syllable string acquisition part (3), according to the pronunciation of each structured word that is divided, generates the syllable string of each structural word; The ellipsis language generation rule storage part (6), Storing the generation rules of abbreviations using syllables; the abbreviation generation unit (7) takes out syllables from the syllable strings of each structural word and connects them to generate candidates for abbreviations composed of one or more syllables, Abbreviations are generated by using abbreviation generation rules for these candidates; and the vocabulary storage unit (8) stores the generated abbreviations together with the recognition target words as the speech recognition dictionary.

Description

technical field [0001] The present invention relates to a voice recognition lexicon creating device for creating a dictionary used in a voice recognition device targeting an unspecified speaker, and a voice recognition device for recognizing a voice using the dictionary. Background technique [0002] Conventionally, in a speech recognition device targeting an unspecified speaker, a speech recognition dictionary for specifying recognition vocabulary was indispensable. When the vocabulary to be recognized can be specified at the time of system design, a pre-compiled dictionary for voice recognition is used. However, if the vocabulary cannot be specified or should be changed dynamically, it is manually input or automatically based on the character string information. Vocabulary for voice recognition is compiled and registered in the dictionary. For example, the voice recognition device in the television program switching device analyzes the character string information includi...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/06G10L15/10
CPCG10L15/06G10L15/187
Inventor 冲本纯幸
Owner PANASONIC INTELLECTUAL PROPERTY CORP OF AMERICA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap