Voice-activated control for electrical device

• US 6,594,630 B1
• Filed: 11/19/1999
• Issued: 07/15/2003
• Est. Priority Date: 11/19/1999
• Status: Expired due to Term

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: Voice Signal Technologies Incorporated (Microsoft Corporation)
• Inventors: Zlokarnik, Igor, Roth, Daniel Lawrence
• Primary Examiner(s): To, Doris H.
• Assistant Examiner(s): Nolan, Daniel A.
• Application Number:
  US09/443,957
• Time in Patent Office: 1,334 Days
• Field of Search: 704/270, 704/275, 704/251-256, 704/241
• US Class Current: 704/256.5
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

An apparatus for voice-activated control of an electrical device comprises a receiving arrangement for receiving audio data generated by user. A vioce recognition arrangement is provided for determining whether the received audio data is a command word for controlling the electrical device. The voice recognition arrangement includes a microprocessor for comparing the received audio data with voice recognition data previously stored in the voice recognition arrangement. The voice recognition arrangment generates at least one control signal based on the comparison when the comparison reaches a predetermined threshold value. A power control controls power delivered to the electrical device. The power control is responsive to at least one control signal generated by the voice recognition arrangement for operating the electrical device in response to the at least one audio command generated by the user. An arrangement for adjusting the predetermined threshold value is provided to cause a control signal to be generated by the voice recognition arrangement when the audio data generated by the user varies from the previously stored voice recognition data.

Claims

1. An apparatus for voice activated control of an electrical device, the apparatus comprising:

receiving means for receiving at least one audio command generated by a user, the at least one audio command having a command word portion and a pause portion, each of the audio command portions being at least one syllable in length;

speech recognition data having a command word portion and a pause portion, each of the speech recognition data portions being at least one syllable in length;

speech recognition means including a Hidden Markov Model for comparing said command word portion and said pause portion of said at least one received audio command with said command word portion and said pause portion, respectively, of said speech recognition data, said speech recognition means generating at least one control signal based on said comparison, said speech recognition means prevents operation of the electrical device when the spectral content is dynamic;

means for analyzing the pause portion of the received audio command for spectral content; and

power control means for controlling power delivered to an electrical device, said power control means being responsive to said at least one control signal generated by said speech recognition means for operating the electrical device in response to said at least one audio command generated by the user.

2. The apparatus of claim 1, wherein said receiving means receives background noise data in conjunction with said audio command, and further comprising means for generating a command word score and a background noise score based on the comparison of the received audio command to the speech recognition data and the background noise data, respectively, said speech recognition means generating said at least one control signal when said command word score exceeds said background noise score.

3. The apparatus of claim 2, and further comprising:

means for analyzing the command word portion of the received audio command and the background noise data for energy content; and

means for comparing the energy content of the command word portion to the energy content of the background noise data and generating a corresponding energy comparison value;

wherein said speech recognition means prevents the generation of said at least one control signal when said energy comparison value is below a predetermined level.

4. The apparatus of claim 1, wherein said receiving means receives background noise data in conjunction with said audio command, and further comprising:

means for analyzing the command word portion of the receive audio command and the background noise data for energy content; and

means for comparing the energy content of the command word portion to the energy content of the background noise data and generating a corresponding energy comparison value;

wherein said speech recognition means prevents the generation of said at least one control signal when said energy comparison value is below a predetermined level.

5. The apparatus of claim 1, wherein each of said at least one audio command and said speech recognition data comprises first and second command word portions separated by said pause portion and further comprising a second pause portion having one syllable in duration before said first command word portion and a third pause portion having one syllable in duration after said second command word portion.

6. The apparatus of claim 1, wherein the speech recognition means further including a microcontroller with a fixed-point embedded microprocessor, the microprocessor is chosen from the group of 8-bit and 16-bit micro controller unit-microprocessors.

7. A method of activating an electrical device through at least one audio command from a user, the method comprising the steps of:

recording speech recognition data having a command word portion and a pause portion, each of the speech recognition data portions being at least one syllable in length;

receiving at least one audio command from a user, the at least one audio command having a command word portion and a pause portion, each of the audio command portions being at least one syllable in length;

comparing said command word portion and said pause portion of said at least one received audio command with said command word portion and said pause portion, respectively, of said speech recognition data;

generating at least one control signal based on said comparison;

controlling power delivered to an electrical device in response to said at least one control signal for operating the electrical device in response to said at least one received audio command;

analyzing the pause portion of the received audio command for spectral content; and

preventing operation of the electrical device when the spectral content is dynamic.

8. The method of claim 7, wherein the step of recording speech recognition data includes recording the voice of a user while the user utters said at least one audio command.

9. The method of claim 8, and further comprising the steps of receiving background noise data in conjunction with receiving said at least one audio command;

comparing the background noise data to the at least one received audio command;

generating a command word score from the comparison of the received audio command to the voice recognition data;

generating a background noise score based on the comparison of the received audio command to the background noise data; and

generating said at least one control signal when said command word score exceeds said background noise score.

10. The method of claim 7, and further comprising:

ascertaining a first energy content for the command word portion of the received audio command;

ascertaining a second energy content for the received background noise data;

comparing the first and second energy contents and generating an energy comparison value; and

preventing the generation of said at least one control signal when said energy comparison value is below a predetermined level.

11. The method of claim 9, wherein the step of generating a command word score includes assigning a word entrance penalty to the command word portion and the pause portion.

12. The method of claim 11, wherein the word entrance penalty of the pause portion is negative.

13. The apparatus of claim 12, and further comprising a second pause portion having one syllable in duration before said first command word portion and a third pause portion having one syllable in duration after a second command word portion.

14. The method of claim 7, and further comprising:

receiving background noise data in conjunction with receiving said at least one audio command;

ascertaining a first energy content for the command word portion of the received audio command;

ascertaining a second energy content for the received background noise data;

comparing the first and second energy contents and generating a corresponding energy comparison value; and

preventing the generation of said at least one control signal when said energy comparison value is below a predetermined level.

15. The apparatus of claim 7, wherein each of said at least one audio command and said speech recognition data comprises at least first and second command word portions separated by said pause portion.

16. A method of activating an electrical device through at least one audio command from a user, the method comprising the steps of:

recording speech recognition data having a command word portion and a pause portion, each of the speech recognition data portions being at least one syllable in length;

receiving at least one audio command from a user, the at least one audio command having first and second command word portions and a first, second and third pause portions, each of the audio command portions being at least one syllable in length, said second pause portion having one syllable in duration before said first command word portion and said third pause portion having one syllable in duration after said second command word portion;

comparing said command word portion and said pause portion of said at least one received audio command with said command word portion and said pause portion, respectively, of said speech recognition data;

generating at least one control signal based on said comparison;

controlling power delivered to an electrical device in response to said at least one control signal for operating the electrical device in response to said at least one received audio command.