Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
SPEECH INPUT METHOD AND RECORDING MEDIUM
Document Type and Number:
Japanese Patent JP2000047688
Kind Code:
A
Abstract:

To automatically insert a punctuation mark into a natural flow of dictation by discriminating a predetermined specific character string from character strings recognized according to an inputted speech and adding one punctuation mark.

When speech input is started (S2), a KANA(Japanese syllabary) character string is discriminated (S3) corresponding to the progress of speech. A clause structure is analyzed from the KANA character string (S4) and KANJI (Chiniese character) automatic conversion is performed by clauses (S5). Then the necessity of punctuation mark addition is analyzed from the character string recognized according to the inputted speech and a punctuation mark is automatically added (S6). In concrete, it is decided that the ending of a clause is a predetermined specific character or character string and one punctuation mark is added. Then it is confirmed whether or not the speech input ends (S7). Consequently, even when the speech input of a document is performed along a natural flow of dictation at the time of the creation of the document, the document is punctuated properly with punctuation marks to facilitate corrections after the speech input.


Inventors:
TANAKA MASAHIDE
Application Number:
JP21213598A
Publication Date:
February 18, 2000
Filing Date:
July 28, 1998
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
NIKON CORP
International Classes:
G10L15/18; G10L15/183; G10L15/22; (IPC1-7): G10L15/22



 
Previous Patent: SPEECH RECOGNITION DEVICE

Next Patent: SPEECH RECOGNITION DEVICE