Data processing: speech signal processing – linguistics – language – Speech signal processing – Recognition
Reexamination Certificate
1997-10-20
2001-03-20
Knepper, David D. (Department: 2748)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Recognition
C704S231000, C704S270000
Reexamination Certificate
active
06205425
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the detection, analysis, and recognition of speech and is especially concerned with systems utilizing speech air flow, and, also concerned with positioning a plurality of transducers in front of a speaker's mouth for detecting and responding to air flow patterns in space and time. The present invention is particularly concerned with systems and methods for recognizing speech utterances by the aerodynamics and acoustics of the speech utterances.
The present invention also relates to adding an aerodynamic dimension with regard to conventional acoustic only speech signals.
2. Description of the Prior Art
Numerous systems for speech detection, analysis, and recognition and speech control have been proposed. See, for example, U.S. Pat. Nos. 4,718,096 issued to Meisel for “SPEECH RECOGNITION SYSTEM”, 4,590,604 issued to Feilchenfeld for “VOICE-RECOGNITION ELEVATOR SECURITY SYSTEM”, and 4,618,985 issued to Pfeiffer for “SPEECH SYNTHESIZER”, and U.K. Patent Application No. 2,087,617 A by Ichikawa et al for “CONTINUOUS SPEECH RECOGNITION METHOD AND APPARATUS”, and French Patent Application No. 2,559,325 by Dubus et al for speech capture apparatus.
SUMMARY OF THE INVENTION
The present invention is directed to a system and method for analysing and recognizing speech by air flow assciated with air flow through the vocal chord and by the acoustic sound produced during speech production.
Accordingly the present invention also provides a system and method for achieving speech commands by the detection of air flow pattern in the proximity of the mouth in space and time and the detection of spoken sound during an utterance.
And accordingly the present invention also provides a system and method for utilizing the aerodynamics and acoustics of speech utterances.
The system and method of the present invention have advantage over the reference patents cited above that it may work with both audible and inaudible utterances.
In addition, existing speech recognition systems have difficulty in recognizing speech utterances which sound alike even though such utterances may have quite different air flow in the proximity of the mouth in space and time. The present invention has the advantage of being capable of identifying different speech sounds that sound alike whereas their air flow patterns are quite different. For example, many speech words that sound alike may begin with say, “b”, “t”, “d”, “p”, “f”, or “th”, respectively, which have different air flow patterns and/or other air flow properties in the space around the mouth when uttered. Of course, different air flow patterns may also be detected and identified in the middle portions of speech words that sound alike to the ear.
The system and method of the present invention also identifies inaudible rush of air out of or into the mouth during an utterance and between utterances.
Another advantage is that since the method of the present invention detects the air flow pattern around the mouth, it may differentiate between noises in the surrounding from the utterance picked up out of the mouth of a speaker.
The present invention has another advantage that it identifies the speech utterance out of the mouth of a speaker physically present in which case air flow pattern is detected and differentiates from speech utterance from a speaker at a distance in which case no air flow pattern is picked up.
An object of the present invention is to detect and utilize speech air flow, for example, for analysing and studying speech, for recognizing speech, and for accomplishing speech commands, etc.
Another object of the present invention is to provide a device for measuring speech air flow.
Another object of the present invention is to analyse speech by air flow properties.
Another object of the present invention is to recognize speech by air flow properties.
Another object of the present invention is to utilize speech air flow properties as speech commands.
Another object of the present invention is to provide a speech recognition system operable at high noise levels.
Another object of the present invention is to provide a speech control system operable at high noise levels.
Another object of the present invention is to provide means for extracting and analysing speech signals comprising analogues of speech air flow.
Another object of the present invention is to provide means for extracting speech features comprising air flow features.
Another object of the present invention is to provide a system wherein speech is defined by quantities comprising quantities representative of speech air flow properties.
Another object of the present invention is to provide a system wherein speech air flow properties are used to time the speed of a speech utterance.
Another object of the present invention is to provide a speech controlled system operable with softly uttered commands, such as whispered commands.
Another object of the present invention is to provide a system for distinguishing speech entered by a human speaker from background sounds.
Another object of the present invention is to provide a system for utilizing the analogues of air flow quantities of speech.
Another object of the present invention is to utilize speech air flow quantities as means for measuring the speed of speech production.
Another object of the present invention is to detect plosive and fricative utterances.
Another object of the present invention is to provide a speech recognition system relatively immune to background noise sounds.
Another object of the present invention is to provide a system for responding to speech inputs, comprising a plurality of transducers mounted in front of a speaker's mouth to respond to air flow patterns with the output of the transducers being connected to a computer system or other responding subsystems.
Another object of the present invention is to provide a system for recognizing the emotion or manner with which speech is input, for example whether the input speech is spoken in a relaxed manner or a stressed manner.
Another object of the present invention is to provide a headgear (e.g. a helmet) comprising a plurality of transducers mounted on the headgear for detecting and untilizing speech air flow from the mouth of a speaker wearing said headgear.
Another object of the present invention is to provide a helmet for speech input by inputting the aerodynamic signal and acoustic signal of speech utterance.
Another object of the present invention is to provide a means for speech input by inputting the aerodynamic signal and acoustic signal of speech utterance which is relatively insensitive to the speaker head movement.
Another object of the present invention is to provide a helmet for speech input which is relatively insensitive to external noise.
Another object of the present invention is to provide a helmet for speech input which is relatively insensitive to external aerodynamic wind.
Further objects and advantages of my invention will become apparent from a consideration of the drawings and ensuing description of it.
The same system and method of the present invention, together with the advantages thereof, may best be understood by reference to the following description taken in connection with the accompanying drawings, in the several figures of which, like reference numerals identify like elements.
REFERENCES:
patent: 4335276 (1982-06-01), Bull et al.
patent: 4718096 (1988-01-01), Meisel
patent: 5680505 (1997-10-01), Ho
LandOfFree
System and method for speech recognition by aerodynamics and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for speech recognition by aerodynamics and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for speech recognition by aerodynamics and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2522715