ADAPTIVE AUDITORY PATTERN RECOGNITION SYSTEM FOR DRIVER TRAINING AND TESTING EQUIPMENT

• US 3,591,931 A
• Filed: 10/30/1969
• Issued: 07/13/1971
• Est. Priority Date: 10/30/1969
• Status: Expired due to Term

• Current Assignee: Donald H. Schuster
• Original Assignee: Donald H. Schuster
• Inventors: Schuster, Donald H.
• Application Number:
  US04/872,695
• Time in Patent Office: 621 Days
• US Class Current: 434/69
• CPC Class Codes: G09B 9/05 the view from a vehicle being simulated G09B9/052, G09B9/058 take precedence

Abstract

A driver trainer simulator involving a movie film for portraying a series of driving situations projected on a screen in view of a driver station on a driver trainer unit, the movie film being encoded with electrical signals which correspond to successive driving situations on the film and are compared with signals received from the driving controls of the driver trainer upon being operated by a student, an instantaneous display panel for indicating the correctness of the response by the student to each of the successive driving situations and a permanent magneticrecording means for storing the comparative information. An adaptive stressor unit having an audio unit may be simultaneously used to provide secondary auditory perceptual loading on the student. An auditory pattern recognition device including an audio pickup may be provided to receive the audible responses from the student in response to the sound patterns from the stressor unit. An indicator is provided for indicating correct and incorrect audible responses as compared with the audible responses from the audio unit of the stressor unit. The recognition device indicator may be coupled to the stressor unit to control the rate of loading of the auditory signals on the student.

Claims

1. A driver simulator, comprising, a movie film for portraying a series of driving situations projected on a screen in view of a driver station on a driver trainer unit, an adaptive stressor unit having an audio unit which provides secondary auditory perceptual loading sound patterns on the student in the driver trainer unit, an auditory pattern recognition device that includes an audio pickup to receive the audible responses from the student in response to the sound patterns from the stressor unit, said auditory recognition device including an indicator for indicating correct and incorrect audible responses from the student as compared with the sound patterns from the audio unit of the stressor unit, said recognition device indicator being coupled to said stressor unit to control the rate of loading on the student, and said auditory recognition device including three adaptive circuit means for registering the energy and length of the spoken word, the frequency of the initial consonants, and the frequency of the initial vowel.

2. The structure of claim 1 wherein said audible responses from the student include only the words zero and one.

3. The structure of claim 2 wherein said auditory recognition device includes a word detector responsive to the sound energy exceeding a predetermined threshold level for determining when a word has started and when it has stopped, a vowel detector for determining when the vowel in the first syllable of the spoken word has started and when it has stopped by recognizing the constant frequency energy of the vowel.

4. The structure of claim 3 wherein the adaptive means for registering the energy and length of the spoken word includes a word length circuit adapted to be turned on in response to a word being spoken and turned off again when the word is finished, said word length circuit having an output when turned on received by an integrator-decider circuit adapted to provide an output whenever the word length pulse has exceeded a predetermined minimum time duration, a memory means receives an output pulse from said integrator-decider corresponding to the long word zero and registers no pulse in response to the short word one, and the output of the memory means is received by a majority logic decision making circuit.

5. The structure of claim 4 wherein the output of the memory means is received by a decision averager adapted to register the number of pulses received from the integrator-decider, said decision averager being enabled by said word detector such that over approximately 1 minute period of time the word and integrator-decider pulses are averaged, upon the average falling below 50 percent zeros a correction bias is applied to the integrator such that the minimum time is lengthened or shortened approximately to bring the percentage of zero decisions to approximately 50 percent.

6. The structure of claim 4 wherein the adaptive means for registering the frequency of the initial consonants includes high and low pass filters adapted to separate energy frequency bands, and a pair of envelope detectors are adapted to receive the output of the high- and low-pass filters which in turn feed a differential amplifier comparator, the output of the comparator being positive or negative depending on whether a high frequency consonant or a low frequency consonant has been spoken, a memory means connected to said comparator and said vowel and word detectors for receiving the output of the comparator during a restricted period of time established by when a spoken digit starts as determined by the start of The word detector pulse, and the start of the vowel in the first syllable as determined by the start of the vowel pulse whereby the restricted period is started by the word pulse and stopped by the vowel pulse and the memory means receives a signal from the comparator only during the initial consonant of a word, and the ratio of the high and low frequency components is remembered for the duration of the word pulse, and the output of the memory means is connected to the majority logic recognition decision circuit.

7. The structure of claim 6 wherein said adaptive means for registering the frequency of the initial vowel includes high- and low-pass filters adapted to separate energy frequency bands, a pair of envelope detectors are adapted to receive the output of the high- and low-pass filters which in turn feed a differential amplifier comparator, the output of the comparator being positive or negative depending on whether a high frequency vowel or a low frequency vowel has been spoken, a memory means connected to said comparator and said vowel detector for receiving the output of the comparator during a restricted period of time established by when a vowel in the first syllable is present as determined by the vowel detector pulse being received by the memory means, and the ratio of the high- and low-frequency components is remembered for the duration of the word pulse;

and the output of the memory means is connected to the majority logic recognition circuit.

8. The structure of claim 7 wherein said majority logic recognition circuit includes an output signal indicating a one or a zero was spoken depending on the output of energy and length adaptive circuit, the frequency of the initial consonant adaptive circuit, and the frequency of the initial vowel adaptive circuit compared with a predetermined output pattern, the outputs of two or more of said adaptive circuits controlling the output of said majority logic recognition circuit.

9. The structure of claim 1 wherein a majority logic recognition circuit is connected to the outputs of said three adaptive circuit means and the output signal thereof indicates a one or zero was spoken depending on the output of said three adaptive circuit means compared with a predetermined output pattern, the outputs of two or more of said three adaptive circuit means controlling the output of said majority logic recognition circuit.