Speech recognition system for an automotive vehicle

• US 4,501,012 A
• Filed: 05/18/1982
• Issued: 02/19/1985
• Est. Priority Date: 11/17/1980
• Status: Expired due to Term

• Current Assignee: Nissan Motor Co., Ltd.
• Original Assignee: Nissan Motor Co., Ltd.
• Inventors: Kishi, Norimasa, Noso, Kazunori
• Primary Examiner(s): Kemeny, E. S. Matt
• Application Number:
  US06/379,506
• Time in Patent Office: 1,008 Days
• Field of Search: 381/41-43, 364/513, 364/513.5, 364/424, 364/426, 364/440, 180/170, 180/179, 180/281, 367/198, 340/52 F
• US Class Current: 704/275
• CPC Class Codes: B60R 16/0373 Voice control in general G10L
• CPC Class Codes: G10L 15/00 Speech recognition G10L17/00 takes precedence
• CPC Class Codes: G10L 25/87 Detection of discrete points within a voice signal

Abstract

A speech recognition system for an automotive vehicle which can selectively activate actuators according to vehicle-travelling conditions in such a way that a spoken instruction such as "Door open" or "Trunk lid open" can be recognized only when vehicle speed is zero, "Headlight high/low beam" can be recognized only when vehicle speed reaches a predetermined value, and "Air conditioner on/off" or "Radio on/off" can be recognized regardless of vehicle speed. The speech recognition system for an automotive vehicle according to the present invention comprises a speed sensor or the like and a speed-dependent speech selection unit including AND gates or analog switches, in addition to the conventional speech recognizer.

Claims

1. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions, which comprises:

(a) at least one microphone for transducing spoken instructions into corresponding electric signals and outputting signals corresponding thereto;

(b) a speech recognizer including a parameter extraction section, at least one reference pattern memory unit, a resemblance comparator, a resemblance determination section, and a controller, said speech recognizer being connected to said at least one microphone for extracting time-series matrix-phonetic pattern data from the transduced spoken instruction signals by said parameter extraction section, sequentially comparing the extracted matrix-phonetic pattern data with the reference matrix-phonetic pattern data of a plurality of reference spoken instructions previously stored in said reference pattern memory unit by said resemblance comparator in order to obtain resemblance rates of the extracted pattern data to the reference pattern data, and outputting a command signal corresponding to the transduced spoken instruction signals to one of the actuators when one of the obtained resemblance rates lies within a predetermined range in said resemblance determination section, all the mutual operations within said speech recognizer being executed in accordance with command signals outputted from said controller,

(c) at least one means for detecting vehicle-travelling conditions and outputting signals corresponding thereto; and

(d) a speed-dependent speech selection unit connected to both said speech recognizer and said vehicle-travelling condition detecting means for selectively allowing said speech recognizer to activate some of the actuators, according to the detected vehicle-travelling conditions, in response to the spoken instructions corresponding to the actuators,

whereby it is possible to prevent actuator operations due to said speech recognizer which would be inappropriate for the current vehicle speed.

2. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in claim 1, wherein said speed-dependent speech selection unit is connected between said controller and a plurality of said reference pattern memory units storing a plurality of reference time-series matrix-phonetic pattern data independently for allowing a command signal to pass from said controller to said respective reference pattern memory units therethrough, according to vehicle-travelling conditions, to transfer the respective reference matrix-phonetic pattern data to said resemblance comparator.

3. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in claim 1, wherein said speed-dependent speech selection unit is connected between said speech recognizer and a plurality of the actuators for allowing a command signal to pass from said speech recognizer to said respective actuator therethrough, according to vehicle-travelling conditions, to activate the respective actuator.

4. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in claim 1, wherein said speed-dependent speech selection unit is connected between said at least one microphone and said speech recognizer for allowing signals transduced by said microphone to pass to said speech recognizer therethrough, according to vehicle-travelling conditions, to activate the respective actuator.

5. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken word instructions as set forth in claim 2, wherein said vehicle-travelling condition detecting means is a speed sensor for generating pulse signals, the frequency of which is proportional to vehicle speed, and said speed-dependent speech selection unit comprises:

(a) a frequency-voltage converter connected to said speed sensor for converting the pulse frequency from said speed sensor into a voltage signal representative of vehicle speed and outputting signals corresponding thereto;

(b) a first comparator, the minus input terminal of which is connected to said f-V converter and to the plus input terminal of which a lower reference voltage indicative of a lower refrence vehicle speed is applied, said first comparator comparing the two input signals and outputting a H-voltage level signal when the voltage signal from said f-V converter drops below the lower reference voltage;

(c) a second comparator, the plus input terminal of which is connected to said f-V converter and to the minus input terminal of which a higher reference voltage indicative of a higher reference vehicle speed is applied, said second comparator comparing the two input signals and outputting a H-voltage level signal when the voltage signal from said f-V converter exceeds the higher reference voltage;

(d) a NOR gate, two input terminals of which are connected to said first and second comparators for NORing the two input signals and outputting a H-voltage level signal when the voltage signal from said f-V converter lies between the higher and lower reference voltages;

(e) said controller being directly connected to said first reference pattern memory unit to transfer the reference matrix-phonetic pattern data in said first reference pattern memory unit to said resemblance comparator, regardless of vehicle speed, said controller outputting a H-voltage level signal when any one of the spoken instructions is inputted to said microphone;

(f) a first AND gate, one input terminal of which is connected to said first comparator and the other input terminal of which is connected to said controller for ANDing the two input signals and outputting a H-voltage level signal to said second reference pattern memory unit to transfer the reference matrix-phonetic pattern data in said second reference pattern memory unit to said resemblance comparator when both said first comparator and said controller output a H-voltage level signal, respectively, that is, when vehicle speed drops below the lower reference vehicle speed and any one of the spoken instructions is inputted to said microphone;

(g) a second AND gate, one input terminal of which is connected to said NOR gate and the other input terminal of which is connected to said controller for ANDing the two input signals and outputting a H-voltage level signal to said third reference pattern memory unit to transfer the reference matrix-phonetic pattern data in said third reference pattern memory unit to said resemblance comparator when both said NOR gate and said controller output a high-voltage level signal, respectively, that is, when vehicle speed lies between the lower and higher reference vehicle speed and any one of the spoken instructions is inputted to said microphone; and

(h) a third AND gate, one input terminal of which is connected to said second comparator and the other input terminal of which is connected to said controller for ANDing the two input signals and outputting a H-voltage level signal to said fourth reference pattern memory unit to transfer the reference matrix-phonetic pattern data in said fourth pattern memory unit to said resemblance comparator when said second comparator and said controller output a H-voltage level signal, respectively, that is, when vehicle speed rises beyond the higher reference vehicle speed and any one of the spoken instructions is inputted to said microphone.

6. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in claim 2, wherein one said vehicle-travelling condition detecting means include an ignition switch sensor for outputting a H-voltage level signal when an ignition switch is turned on, a drive position sensor for outputting a H-voltage level signal when a transmission system is shifted into any of 1st DRIVE, 2nd DRIVE and REVERSE positions, and a neutral position sensor for outputting a H-voltage level signal when a transmission system is shifted into NEUTRAL position, and said speed-dependent speech selection unit comprises:

(a) an inverter, input terminal of which is connected to said ignition switch sensor for inverting the input signal and outputting a H-voltage level signal when the ignition switch is turned off;

(b) a fourth AND gate, one input terminal of which is connected to said ignition switch sensor and the other input terminal of which is connected to said neutral position sensor for ANDing the two input signals and for outputting a H-voltage level signal when the ignition switch is turned on and the transmission system is shifted into the NEUTRAL position;

(c) a fifth AND gate, one input terminal of which is connected to said ignition switch sensor and the other input terminal of which is connected to said drive position sensor for ANDing the two input signals and outputting a H-voltage level signal when the ignition switch is turned on and the transmission system is shifted into any of 1st DRIVE, 2nd DRIVE and REVERSE positions;

(d) said controller being directly connected to said first reference pattern memory unit to transfer the reference matrix-phonetic pattern data in said first reference pattern memory unit to said resemblance comparator, regardless of vehicle speed, said controller outputting a H-voltage level signal when any one of the spoken instructions is inputted to said microphone;

(e) a first AND gate, one input terminal of which is connected to said inverter and the other input terminal of which is to said controller for ANDing the two input signals and outputting a H-voltage level signal to said second reference pattern memory unit to transfer the reference matrix-phonetic pattern data in said second reference pattern memory unit to said resemblance comparator when said inverter and said controller output a H-voltage level signal, respectively, that is, when the ignition switch is off and any of the spoken instructions is inputted to said microphone;

(f) a second AND gate, one input terminal of which is connected to said fourth AND gate and the other input terminal of which is connected to said controller for ANDing the two input signals and outputting a H-voltage level signal to said third reference pattern memory unit to transfer the reference matrix-phonetic pattern data in said third reference pattern memory unit to said resemblance comparator when the ignition switch is turned on, the transmission system is shifted into NEUTRAL position and any one of the spoken instructions is inputted to said microphone; and

(g) a third AND gate, one input terminal of which is connected to said fifth AND gate and the other input terminal of which is connected to said drive position sensor for ANDing the two input signals and outputting a H-voltage level signal to said fourth reference pattern memory unit to transfer the reference matrix-phonetic pattern data in said fourth pattern memory unit to said resemblance comparator when said ignition switch is turned on, the transmission system is shifted into any of 1st DRIVE, 2nd DRIVE and REVERSE positions, and any one of the spoken instructions is inputted to said microphone.

7. A speed recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in claim 3, wherein said vehicle-travelling condition detecting means is a speed sensor for generating pulse signals, the frequency of which is proportional to vehicle speed and said speed-dependent speech selection unit comprises:

(a) a frequency-voltage converter connected to said speed sensor for converting the pulse frequency from said speed sensor into a voltage signal representative of vehicle speed and outputting signals corresponding thereto;

(b) a comparator the minus input terminal of which is connected to said f-V converter and to the plus input terminal of which a reference voltage indicative of a reference vehicle speed is applied, said comparator comparing the two input signals thereof and outputting a H-voltage level signal when the voltage signal from said f-V converter drops below the reference voltage;

(c) a transistor the base terminal of which is connected to said comparator and emitter terminal of which is grounded for inverting the voltage signal from said comparator at the collector terminal thereof;

(d) said speech recognizer being directly connected to a first actuator for activating said first actuator, regardless of vehicle speed, when said speech recognizer outputs a H-voltage level signal, that is, when a spoken instruction corresponding to said first actuator is inputted to said microphone;

(e) a first AND gate, one input terminal of which is connected to said comparator and the other input terminal of which is connected to said speed recognizer for ANDing the two input signals and for outputting a signal to said second actuator to activate said second actuator when both said comparator and said speech recognizer output a H-voltage level signal, respectively, that is, when vehicle speed drops below the reference vehicle speed and a spoken instruction corresponding to said second actuator is inputted to said microphone; and

(f) a third AND gate, one input terminal of which is connected to said speed recognizer and the other input terminal of which is connected to the collector terminal of said transistor for ANDing the two input signals thereof and outputting a H-voltage level signal to said forth actuator to activate said fourth actuator when both said transistor and said speech recognizer output a H-voltage level signal, respectively, that is, when vehicle speed rises beyond the reference vehicle speed and a spoken instruction corresponding to said fourth actuator is inputted to said microphone.

8. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken word instructions as set forth in claim 3, wherein said vehicle travelling condition detecting means include an ignition switch sensor for outputting a H-voltage level signal when an ignition switch is turned on and a drive position sensor for outputting a H-voltage level signal when a transmission system is in any of positions 1st DRIVE, 2nd DRIVE and REVERSE positions, and said speed-dependent speech selection unit comprises:

(a) an inverter, the input terminal of which is connected to said ignition switch for inverting the input signal thereof and outputting a H-voltage level signal when the ignition switch is turned off;

(b) a fifth AND gate, one input terminal of which is connected to said ignition switch sensor and the other input terminal of which is connected to said drive position sensor for ANDing the two input signals thereof and outputting a H-voltage level signal when the ignition switch is turned on and the transmission system is shifted into any of 1st DRIVE, 2nd DRIVE and REVERSE positions;

(c) said speech recognizer being directly connected to a first actuator for activating said first actuator, regardless of vehicle speed, when said speech recognizer outputs a H-voltage level signals, that is, when a spoken instruction corresponding to said first actuator is inputted to said microphone;

(d) a first AND gate, one input terminal of which is connected to said inverter and the other input terminal of which is connected to said speech recognizer for ANDing the two input signals and for outputting a signal to said second actuator to activate said second actuator when said inverter and said speech recognizer output a H-voltage level signal, respectively, that is, when vehicle speed is zero and a spoken instruction corresponding to said second actuator is inputted to said microphone; and

(e) a third AND gate, one input terminal of which is connected to said speech recognizer and the other input terminal of which is connected to said fifth AND gate for ANDing the two input signals thereof and outputting a H-voltage level signal to said fourth actuator to activate said fourth actuator when both said fifth AND gate and said speech recognizer outputs a H-voltage level signal, respectively, that is, when vehicle speed is not zero and a spoken instruction corresponding to said fourth actuator is inputted to said microphone.

9. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken word instructions as set forth in claim 4, wherein said vehicle-travelling condition detecting means is a speed sensor for generating pulse signals the frequency of which is proportional to vehicle speed and said speed-dependent speech selection unit comprises:

(a) a frequency-voltage converter connected to said speed sensor for converting the pulse frequency from said speed sensor into a voltage signal representative of vehicle speed and outputting signals corresponding thereto;

(b) a comparator, the minus input terminal of which is connected to said f-V converter and to the plus input terminal of which a reference voltage indicative of a reference vehicle speed is applied, said comparator comparing the two input signals thereof and outputting a H-voltage level signal when the voltage signal from said f-V converter drops belows the reference voltage;

(c) a transistor the base terminal of which is connected to said comparator and the emitter terminal of which is grounded for inverting the voltage signal from said comparator at the collector terminal thereof;

(d) a first microphone being directly connected to said speech recognizer to transmit spoken instruction electric signals transduced by said first microphone to said speech recognizer regardless of vehicle speed;

(e) a first analog switch connected between a second microphone and said speech recognizer, the control terminal of which is connected to said comparator, said first analog switch being turned on to transmit spoken instruction electric signals transduced by said second microphone to said speech recognizer in response to a H-voltage level signal from said comparator, that is, when vehicle speed drops below the reference vehicle speed; and

(f) a third analog switch connected between a third microphone and said speech recognizer, the control terminal of which is connected to the collector terminal of said transistor, said third analog switch being turned on to transmit spoken instruction electric signals transduced by said third microphone to said speech recognizer in response to a H-voltage level signal from said transistor, that is, when vehicle speed rises beyond the reference vehicle speed.

10. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken word instructions as set forth in claim 4, wherein said vehicle travelling condition detecting means include an ignition switch sensor for outputting a H-voltage level signal when an ignition switch is turned on and a drive position sensor for outputting a H-voltage level signal when a transmission system is shifted to any of 1st DRIVE, 2nd DRIVE and REVERSE positions, and said speed-dependent speech selection unit comprises:

(a) an inverter, the input terminal of which is connected to said ignition switch for inverting the input signal thereof and outputting a H-voltage level signal when the ignition switch is turned off;

(b) a fifth AND gate, one input terminal of which is connected to said ignition switch sensor and the other input terminal of which is connected to said drive position sensor for ANDing the two input signals thereof and outputting a H-voltage level signal when the ignition switch is turned on and the transmission system is shifted into any of 1st DRIVE, 2nd DRIVE and REVERSE positions;

(c) a first microphone being directly connected to said speech recognizer to transmit spoken instruction electric signals transduced by said first microphone to said speech recognizer regardless of vehicle speed;

(d) a first analog switch connected between a second microphone and said speech recognizer, the control terminal of which is connected to said inverter, said first analog switch being turned on to transmit spoken instruction electric signals transduced by said second microphone to said speech recognizer in response to a H-voltage level signal from said inverter, that is, when vehicle speed is zero; and

(e) a third analog switch connected between a third microphone and said speech recognizer, the control terminal of which is connected to the collector terminal of said transistor, said third analog switch being turned on to transmit spoken instruction electric signals transduced by said third microphone to said speech recognizer in response to a H-voltage level signal from said transistor, that is, when vehicle speed is not zero.

11. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in either claim 5 or 6, wherein said first reference pattern memory unit stores reference matrix-phonetic pattern data for spoken instructions of "Air conditioner on/off", "Radio on/off", "Channel 1, 2, . . . ", and "Window open/close".

12. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in either claim 5 or 6, wherein said second reference pattern memory unit stores reference matrix-phonetic pattern data for spoken instructions of "Door open", and "Trunk lid open".

13. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in either claim 5 or 6, wherein said third reference pattern memory unit stores reference matrix-phonetic pattern data for spoken instruction of "Headlight on/off".

14. A speech recognition system for an automotive vehicle for outputting a plurality of command signals to a plurality of vehicle actuators independently in accordance with spoken instructions as set forth in either claim 5 or 6, wherein said fourth reference pattern memory unit stores reference matrix-phonetic pattern data for spoken instruction of "Headlight high/low beam".