Embedded system for construction of small footprint speech recognition with user-definable constraints

• US 9,117,449 B2
• Filed: 04/26/2012
• Issued: 08/25/2015
• Est. Priority Date: 04/26/2012
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Newman, Michael Jack, Roth, Robert, Alexander, William D., van Mulbregt, Paul
• Primary Examiner(s): Albertalli, Brian
• Application Number:
  US13/456,959
• Publication Number:
  US 20130289994A1
• Time in Patent Office: 1,216 Days
• Field of Search: 704/254, 704/275, 704/251
• US Class Current: 1/1
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: H04M 1/724 User interfaces specially adapted for cordless or mobile telephones
• CPC Class Codes: H04M 2250/74 with voice recognition means voice activated dialling H04M1/271; voice recognition algorithms G10L15/00
• CPC Class Codes: Y02D 30/70 in wireless communication networks

Abstract

Techniques disclosed herein include systems and methods that enable a voice trigger that wakes-up an electronic device or causes the device to make additional voice commands active, without manual initiation of voice command functionality. In addition, such a voice trigger is dynamically programmable or customizable. A speaker can program or designate a particular phrase as the voice trigger. In general, techniques herein execute a voice-activated wake-up system that operates on a digital signal processor (DSP) or other low-power, secondary processing unit of an electronic device instead of running on a central processing unit (CPU). A speech recognition manager runs two speech recognition systems on an electronic device. The CPU dynamically creates a compact speech system for the DSP. Such a compact system can be continuously run during a standby mode, without quickly exhausting a battery supply.

Claims

1. A computer-implemented method for managing speech recognition, the computer implemented method comprising:

receiving configuration input at a voice-activated wake-up function of an electronic device, the configuration input including a trigger phrase, the configuration input being received at a first processor of the electronic device, the electronic device having a second processor in addition to the first processor, wherein the configuration input and at least one decoy word is first received at the second processor;

creating a finite state transducer network including a network of speech recognition states corresponding to the trigger phrase, the network of speech recognition states being created at the first processor using a first speech recognition engine that the first processor executes, wherein the trigger phrase is evaluated using a recognition grammar without applying a vocabulary model;

transferring the network of speech recognition states from the first processor to the second processor; and

executing a second speech recognition engine on the second processor using the network of speech recognition states corresponding to the trigger phrase, the second processor executing the second speech recognition engine while the first speech recognition engine of the first processor is in an inactive state, wherein the second speech recognition engine includes a finite state transducer decoder configured to execute the finite state transducer network;

wherein the first processor is a central processing unit, and wherein the second processor is a digital signal processor and wherein the first processor and the second processor are co-located within a mobile telephone.

2. The computer-implemented method of claim 1, wherein executing the second speech recognition engine includes:

continuously analyzing spoken utterances using the second processor while the first speech recognition engine is in an inactive state; and

in response to identifying that a particular spoken utterance includes the trigger phrase, signaling the first processor to return the first speech recognition engine to an active state.

3. The computer-implemented method of claim 2, wherein returning the first speech recognition engine to an active state includes the first processor responding to subsequent voice commands.

4. The computer-implemented method of claim 1, wherein executing the second speech recognition engine while the first speech recognition engine of the first processor is in an inactive state includes the first processor being in an inactive voice command mode;

and

in response to identifying that a particular spoken utterance includes the trigger phrase, causing the electronic device to switch from an inactive voice command mode to an active voice command mode.

5. The computer-implemented method of claim 1, wherein the second processor executes the second speech recognition engine providing the voice-activated wake-up function while the electronic device is in a standby mode.

6. The computer-implemented method of claim 1, wherein transferring the speech recognition states to the second processor includes transferring the speech recognition states to a digital signal processor (DSP).

7. The computer-implemented method of claim 1, wherein receiving the configuration input includes receiving the trigger phrase as text input via a user interface of the electronic device.

8. The computer-implemented method of claim 7, wherein receiving the configuration input includes receiving a spoken utterance that confirms the text input.

9. The computer-implemented method of claim 1, wherein the second processor uses less power compared to the first processor when the first processor is in an active state.

10. A system for managing speech recognition, the system comprising:

a first processor;

a second processor; and

a memory coupled to the processors, the memory storing instructions that, when executed by the processors, cause the system to perform the operations of;

receiving configuration input at a voice-activated wake-up function of an electronic device, the configuration input including a trigger phrase, the configuration input being received at a first processor of the electronic device, the electronic device having a second processor in addition to the first processor, wherein the configuration input and at least one decoy word is first received at the second processor;

creating a finite state transducer network including a network of speech recognition states corresponding to the trigger phrase, the network of speech recognition states being created at the first processor using a first speech recognition engine that the first processor executes, wherein the trigger phrase is evaluated using a recognition grammar without applying a vocabulary model;

transferring the network of speech recognition states from the first processor to the second processor; and

executing a second speech recognition engine on the second processor using the network of speech recognition states corresponding to the trigger phrase, the second processor executing the second speech recognition engine while the first speech recognition engine of the first processor is in an inactive state, wherein the second speech recognition engine includes a finite state transducer decoder configured to execute the finite state transducer network;

wherein the first processor is a central processing unit, and wherein the second processor is a digital signal processor and wherein the first processor and the second processor are co-located within a mobile telephone.

11. The system of claim 10, wherein executing the second speech recognition engine includes:

continuously analyzing spoken utterances using the second processor while the first speech recognition engine is in an inactive state; and

in response to identifying that a particular spoken utterance includes the trigger phrase, signaling the first processor to return the first speech recognition engine to an active state.

12. The system claim 11, wherein returning the first speech recognition engine to an active state includes the first processor responding to subsequent voice commands.

13. The system of claim 10, wherein executing the second speech recognition engine while the first speech recognition engine of the first processor is in an inactive state includes the first processor being in an inactive voice command mode;

and

in response to identifying that a particular spoken utterance includes the trigger phrase, causing the electronic device to switch from an inactive voice command mode to an active voice command mode.

14. The system of claim 10, wherein the second processor executes the second speech recognition engine providing the voice-activated wake-up function while the electronic device is in a standby mode.

15. The system of claim 10, wherein transferring the speech recognition states to the second processor includes transferring the speech recognition states to a digital signal processor (DSP).

16. The system of claim 10, wherein receiving the configuration input includes receiving the trigger phrase as text input via a user interface of the electronic device.

17. A computer program product including a non-transitory computer-storage medium having instructions stored thereon for processing data information, such that the instructions, when carried out by a processing device, cause the processing device to perform the operations of:

receiving configuration input at a voice-activated wake-up function of an electronic device, the configuration input including a trigger phrase, the configuration input being received at a first processor of the electronic device, the electronic device having a second processor in addition to the first processor, wherein the configuration input and at least one decoy word is first received at the second processor;

creating a finite state transducer network including a network of speech recognition states corresponding to the trigger phrase, the network of speech recognition states being created at the first processor using a first speech recognition engine that the first processor executes, wherein the trigger phrase is evaluated using a recognition grammar without applying a vocabulary model;

transferring the network of speech recognition states from the first processor to the second processor; and

executing a second speech recognition engine on the second processor using the network of speech recognition states corresponding to the trigger phrase, the second processor executing the second speech recognition engine while the first speech recognition engine of the first processor is in an inactive state, wherein the second speech recognition engine includes a finite state transducer decoder configured to execute the finite state transducer network;

wherein the first processor is a central processing unit, and wherein the second processor is a digital signal processor and wherein the first processor and the second processor are co-located within a mobile telephone.

18. The computer program product of claim 17, wherein executing the second speech recognition engine includes:

continuously analyzing spoken utterances using the second processor while the first speech recognition engine is in an inactive state; and

in response to identifying that a particular spoken utterance includes the trigger phrase, signaling the first processor to return the first speech recognition engine to an active state.

19. A computer-implemented method for managing speech recognition, the computer implemented method comprising:

receiving configuration input at a voice-activated wake-up function of an electronic device, the configuration input including a trigger phrase, the electronic device having a first processor that executes a first speech recognition engine, the electronic device having a second processor in addition to the first processor, wherein the configuration input and at least one decoy word is first received at the second processor;

transmitting the configuration input to a remote server computer;

creating a finite state transducer network including a network of speech recognition states corresponding to the trigger phrase, the network of speech recognition states being created at the remote server computer, wherein the trigger phrase is evaluated using a recognition grammar without applying a vocabulary model;

receiving the network of speech recognition states corresponding to the trigger phrase at the electronic device;

transferring the network of speech recognition states to the second processor; and

executing a second speech recognition engine on the second processor using the network of speech recognition states corresponding to the trigger phrase, the second processor executing the second speech recognition engine while the first speech recognition engine of the first processor is in an inactive state, wherein the second speech recognition engine includes a finite state transducer decoder configured to execute the finite state transducer network;

wherein the first processor is a central processing unit, and wherein the second processor is a digital signal processor and wherein the first processor and the second processor are co-located within a mobile telephone.

20. The computer-implemented method of claim 19, wherein executing the second speech recognition engine includes:

continuously analyzing spoken utterances using the second processor while the first speech recognition engine is in an inactive state; and

in response to identifying that a particular spoken utterance includes the trigger phrase, signaling the first processor to return the first speech recognition engine to an active state.

21. The computer-implemented method of claim 20, wherein returning the first speech recognition engine to an active state includes the first processor responding to subsequent voice commands.

22. The computer-implemented method of claim 19, wherein executing the second speech recognition engine while the first speech recognition engine of the first processor is in an inactive state includes the first processor being in an inactive voice command mode;

and

in response to identifying that a particular spoken utterance includes the trigger phrase, causing the electronic device to switch from an inactive voice command mode to an active voice command mode.