Speech recognition system with an accurate recognition function

• US 5,182,765 A
• Filed: 01/24/1992
• Issued: 01/26/1993
• Est. Priority Date: 11/26/1985
• Status: Expired due to Term

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Ishii, Takaaki, Kuge, Toru
• Primary Examiner(s): BROWN, THOMAS
• Application Number:
  US07/825,421
• Time in Patent Office: 368 Days
• Field of Search: 379/88, 379/58, 379/63, 379/354, 379/355, 379/216, 381/42, 381/43
• US Class Current: 379/88.04
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: H04M 1/271 controlled by voice recognition

Abstract

A speech registration/recognition processing section is responsive to a speech input signal to be registered or recognized from a microphone to selectively subject the speech input signal to a registration or a recognition processing, in which upon the registration processing the speech input signal is allowed to be stored as recognition data and upon the recognition processing the speech input signal is compared to the recognition data stored. A speech record/reproduction processing section is responsive to the speech input signal from the microphone to subject the speech input signal to a record/reproduction processing, in which upon the record processing the speech input signal is recorded as a record signal and upon the reproduction processing the record signal is delivered as a reproduction signal. A speaker is supplied with the reproduction signal and the recognized information is provided to the user through this speaker. A control section is provided, in accordance with a registration or a recognition mode designation signal, for setting the speech registration/recognition processing section to the registration or the recognition processing and for setting the speech record/reproduction processing section to the record or the reproduction processing corresponding to the registration or the recognition processing at which the speech registration/recognition processing section is placed.

Claims

1. A speech recognition system comprising:

microphone means for producing speech input signals of various users, to be registered or recognized;

speech recognition and registration means responsive to a speech input signal from the microphone means, so as to subject the speech input signal to either a registration or a recognition processing, in which upon the registration processing, the speech input signal is allowed to be stored as recognition data and upon the recognition processing, the speech input signal is compared to the recognition data which has been stored; and

a central processing unit which performs speech recognition processing based on the results of comparison;

wherein said speech recognition and registration means comprises a plurality of speech recognition circuits which are connected between a corresponding number of variable characteristic circuits and speech registration RAM'"'"'s respectively where a plurality of modified speech patterns are stored for each word or each clause, said modified speech patterns including, as the characteristics of the speech, the magnitudes of sound volumes, the varying lengths of sounds, the variation of sound intervals, the husky or somewhat masked voices, and

each of said variable characteristic circuits is comprised of an amplifier of a different amplification factor and different bandpath characteristic, a set of resistors and a mode designating switch, respectively, to allow the speech which is input through said microphone means in a speech registration mode to be prepared as a variation pattern so that it is stored as such in the speech registration RAM of the corresponding speech recognition circuit, such that when a user inputs his speech through said microphone means it is supplied respectively through said variable characteristic circuits to said speech recognition circuits where the input speeches are compared to the speech patterns which have been registered in the respective speech registration RAMs.