Method for improving the reliability of voice controls of function elements and device for carrying out this method

• US 4,901,354 A
• Filed: 12/05/1988
• Issued: 02/13/1990
• Est. Priority Date: 12/18/1987
• Status: Expired due to Fees

• Current Assignee: Daimler-Benz AG (Mercedes-Benz Group AG)
• Original Assignee: Daimler-Benz AG (Mercedes-Benz Group AG)
• Inventors: Gollmar, Klaus, Junemann, Gerhard
• Primary Examiner(s): Isen, Forester W.
• Application Number:
  US07/279,814
• Time in Patent Office: 435 Days
• Field of Search: 381/97, 381/110, 381/71, 381/169, 381/94, 381/46
• US Class Current: 381/110
• CPC Class Codes: G10L 15/00 Speech recognition G10L17/00 takes precedence

Abstract

In a method and a device for reliably detecting at least the beginning of a voice command for a voice control of function elements, signals of a first microphone to which contact sound of an operating person is applied, are used for triggering a second microphone which is directed towards the mouth of the operating person in order to improve the reliability of the voice control in high ambient noise.

Claims

1. Method for improving the reliability of voice control systems of function elements at work stations or in motor vehicles, particularly by reliably detecting the beginning of voice commands in a voice recognition circuit, in which mainly sound generated during speaking is applied to a first microphone and at least ambient noises are applied;

to a second microphone and in which misinterpretation of a pure ambient noise as voice command is prevented by signals generated in both microphones being electrically brought into a particular relation with respect to one another, wherein contact measurable sound generated during the speaking of an operating person is applied to the first microphone by direct contact and that the second microphone, which is directed towards the mouth of the operating person with a spatial distance and to which, as a result, sound generated during speaking is also applied, is electrically switched to the voice recognition circuit for the transmission of the voice signal only when the first microphone in turn generates signals from the contact measurable sound.

2. A voice control system including a first microphone to which mainly sound waves generated during speaking can be applied, a second microphone to which at least ambient noises can be applied, and an electric circuit which produces a particular relation between electric signals generated by the two microphones and supplies output signals to a voice circuit, recognition circuit, wherein the first microphone is constructed as a direct-contact microphone resting against a part of the body of an operating person;

the second microphone is directed towards the mouth of the operating person at a spatial distance and is furthermore supplied with sound generated during speaking; and

the electric circuit is constructed as a trigger circuit , which can be actuated by signals generated by the first microphone, for producing a switching relation between signals of the first and the second microphone, which switching relation enables signals of at least the second microphone to be supplied as output signals to the voice recognition circuit only if signals of the first microphone are present at the same time as the signal of the second microphone at the trigger circuit.

3. Device according to claim 2, wherein the first microphone is a throat microphone which can be applied to the neck of the operating person.

4. Device according to claim 3, wherein both microphones are attached to harnesses which can be worn by the operating person.

5. Device according to claim 2, including a timing means with a time constant which is adjustable to the maximum word reception duration of the voice recognition circuit for interrupting the supply of output signals of the trigger circuit to the voice recognition circuit after the time constant set has elapsed.

6. Device according to claim 2, including means for electrically combining the signals of the first and second microphones for optimizing the recognizability of the signals of the second microphone.

7. Device according to claim 2, wherein the trigger circuit includes a threshold switch which compares the electric signals of the first microphone with an adjustable threshold value and a second switch , which can be closed by the threshold switch if the signals of the first microphone are above the threshold value, in series with the path of the electric signals of the second microphone.

8. Device according to claim 7, including a timing means integrated in the second switch, said timing means having a time constant which is adjustable to the maximum word reception duration of the voice recognition circuit for interrupting the supply of output signals of the trigger circuit to the voice recognition circuit after the time constant set has elapsed.

9. A voice control system including a first microphone to which mainly sound waves generated during speaking can be applied, a second microphone to which at least ambient noises can be applied, and an electric circuit which produces a particular relation between electric signals generated by the two microphones and supplies output signals to a voice recognition circuit, wherein:

the first microphone is constructed as a direct-contact microphone resting against a part of the body of an operating person;

the second microphone is directed towards the mouth of the operating person at a spatial distance and is furthermore supplied with sound generated during speaking;

the electric circuit is constructed as a trigger circuit, which can be actuated by signals generated by the first microphone, for producing a switching relation between signals of the first and the second microphone, which switching relation enables signals of at least the second microphone to be supplied as output signals to the voice recognition circuit only if signals of the first microphone are present at the same time as the signal of the second microphone at the trigger circuit; and

a housing, to be carried by the operating person, accommodating the trigger circuit which is electrically connected to both microphones, a transmitter for the wireless transmission of outputs signals of the trigger circuit to a receiver of the voice recognition circuit, and an electric energy source for energizing at least the trigger circuit and the transmitter.

10. Device according to claim 9, wherein the housing includes a first housing part for accommodating the trigger circuit and a second housing part for accommodating the transmitter and the energy source;

and wherein the trigger circuit is electrically connected to the transmitter and the energy source by a multi-pin plug-in connection arranged between and coupling the two housing parts.

11. A voice control system including a first microphone to which mainly sound waves generated during speaking can be applied, a second microphone to which at least ambient noises can be applied, and an electric circuit which produces a particular relation between electric signals generated by the two microphones and supplies output signals to a voice recognition circuit, wherein:

the first microphone is constructed as a direct-contact microphone resting against a part of the body of an operating person;

the second microphone is directed towards the mouth of the operating person at a spatial distance and is furthermore supplied with sound generated during speaking;

the electric circuit is constructed as a trigger circuit, which can be actuated by signals generated by the first microphone, for producing a switching relation between signals of the first and the second microphone, which switching relation enables signals of at least the second microphone to be supplied as output signals to the voice recognition circuit only if signals of the first microphone are present at the same time as the signal of the second microphone at the trigger circuit;

a timing means, with a time constant which is adjustable to the maximum word reception duration of the voice recognition circuit, for interrupting the supply of output signals of the trigger circuit to the voice recognition circuit after the time constant set has elapsed; and

a signal transmitter which is activated for generating a signal to the operating person after the time constant has elapsed.

12. A voice control system including a first microphone to which mainly sound waves generated during speaking can be applied, a second microphone to which at least ambient noises can be applied, and an electric circuit which produces a particular relation between electric signals generated by the two microphones and supplies output signals to a voice recognition circuit, wherein:

the first microphone is constructed as a direct-contact microphone resting against a part of the body of an operating person;

the second microphone is directed towards the mouth of the operating person at a spatial distance and is furthermore supplied with sound generated during speaking;

the electric circuit is constructed as a trigger circuit, which can be actuated by signals generated by the first microphone, for producing a switching relation between signals of the first and the second microphone, which switching relation enables signals of at least the second microphone to be supplied as output signals to the voice recognition circuit only if signals of the first microphone are present at the same time as the signal of the second microphone at the trigger circuit; and

an interrupting switch, to be manually operated, is in series between the first microphone and the trigger circuit.