Speaker verification in a health monitoring system

• US 8,818,810 B2
• Filed: 12/29/2011
• Issued: 08/26/2014
• Est. Priority Date: 12/29/2011
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Weng, Fuliang, Hasan, Taufiq, Feng, Zhe
• Primary Examiner(s): Abebe, Daniel D
• Application Number:
  US13/340,213
• Publication Number:
  US 20130173268A1
• Time in Patent Office: 971 Days
• Field of Search: 704/246, 704/250, 379/88.02
• US Class Current: 704/246
• CPC Class Codes: G10L 17/10 Multimodal systems, i.e. based on the integration of multiple recognition engines or fusion of expert systems
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase
• CPC Class Codes: G16H 40/20 for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms

Abstract

A method for verifying that a person is registered to use a telemedical device includes identifying an unprompted trigger phrase in words spoken by a person and received by the telemedical device. The telemedical device prompts the person to state a name of a registered user and optionally prompts the person to state health tips for the person. The telemedical device verifies that the person is the registered user using utterance data generated from the unprompted trigger phrase, name of the registered user, and health tips.

Claims

1. A method for verifying an identity of a person:

generating, with an audio input device, audio data corresponding to utterances spoken by a person;

identifying first utterance data in the audio data with an audio data processor;

generating an output with a user interface device to prompt the person to speak a registration name in response to the identified first utterance data corresponding to a predetermined trigger utterance;

storing the identified first utterance data in a memory in response to the identified first utterance data corresponding to the predetermined trigger utterance;

generating audio data corresponding to the spoken registration name with the audio input device;

identifying second utterance data in the audio data corresponding to the spoken registration name with the audio data processor;

storing the identified second utterance data in the memory;

verifying, with a speaker verification module, that the person is a user registered in a registration database in association with the registration name in response to the first and second utterance data stored in the memory corresponding to a predetermined model of a voice of the user registered in the registration database in association with the registration name; and

generating an output with the user interface device to offer services to the person in response to the speaker verification module verifying that the person is the user registered in the registration database.

2. The method of claim 1 further comprising:

generating an output with the user interface device to prompt the person to speak a predetermined phrase in response to the speaker verification module identifying that the first utterance data and second utterance data in the memory are insufficient to verify the person with the predetermined model of the voice of the user;

generating audio data corresponding to the spoken predetermined phrase with the audio input device;

identifying third utterance data in the audio data corresponding to the spoken predetermined phrase with the audio data processor;

storing the third utterance data in the memory; and

verifying, with the speaker verification module, that the person is the user registered in the registration database in response the first, second, and third utterance data stored in the memory corresponding to a predetermined model of a voice of the user registered in the registration database.

3. The method of claim 2 further comprising:

storing the third utterance data in the memory in response to the third utterance data corresponding to predetermined utterance data of the predetermined phrase.

4. The method of claim 2 further comprising:

generating the output with the user interface device to prompt the person to speak the predetermined phrase a second time in response to the third utterance data not corresponding to predetermined utterance data of the predetermined phrase.

5. The method of claim 2, the user interface device generating a prompt for the person to speak a health tip as the predetermined phrase.

6. The method of claim 5, the user interface device generating a prompt for the person to speak a health tip that is associated with the registration name in the registration database.

7. The method of claim 2 further comprising:

verifying, with the speaker verification module, that the person speaking the registration name is not the user registered in the registration database in association with the registration name in response to the first, second, and third utterance data stored in the memory not corresponding to the predetermined model of the voice of the user registered in the registration database; and

generating an output with the user interface device to deny services to the person in response to the speaker verification module verifying that the person speaking the registration name is not the user registered in the registration database.

8. The method of claim 2 further comprising:

continuing to generate an output with the user interface device to prompt the person to speak at least one additional predetermined phrase in response to the speaker verification module identifying that the first, second, and third utterance data in the memory are insufficient to verify the person with the predetermined model of the voice of the user;

generating audio data corresponding to the at least one additional predetermined phrase spoken by the person with the audio input device;

identifying at least one additional utterance data in the audio data corresponding to the at least one additional predetermined phrase with the audio data processor;

storing the at least one additional utterance data in the memory; and

verifying, with the speaker verification module, that the person is the user registered in the registration database in association with the registration name in response the first, second, third, and at least one additional utterance data stored in the memory corresponding to the predetermined model of the voice of the user registered in the registration database.

9. The method of claim 8 further comprising:

identifying that the speaker verification module is unable to verify that the person is the user associated with the registration name in the registration database in response to the memory having insufficient utterance data for the speaker verification module to verify that the person is the user after a number of additional utterance data exceeding a predetermined threshold are stored in the memory.

10. The method of claim 9 further comprising:

prompting the person, with the user interface device, to input information for verification of the person with a user input device that is different from the audio input device in response to the speaker verification module being unable to verify that the person is the user associated with the registration name in the registration database.

11. A telemedical device with speaker verification comprising:

an audio input device configured to generate audio data from utterances spoken by a person;

an audio data processor operatively connected to the audio input device and configured to generate utterance data from audio data generated by the audio input device;

a memory configured to store a plurality of utterance data generated by the audio data processor;

a registration database configured to associate at least one user with a registration name and a voice model corresponding to the at least one user;

a speaker verification module operatively connected to the memory and the registration database;

a user interface device; and

a controller operatively connected to the audio input device, audio data processor, memory, registration database, speaker verification module, and user interface device, the controller being configured to;

activate the audio input device to receive sounds including utterances spoken by a person and to generate audio data corresponding to the utterances without prompting the person to speak;

identify first utterance data in audio data corresponding to the utterances spoken by the person with the audio data processor;

store the identified first utterance data in the memory;

generate an output with the user interface device to prompt the person to speak a registration name in response to the first utterance data corresponding to a predetermined trigger utterance;

generate audio data corresponding to the spoken registration name with the audio input device;

identify second utterance data in the audio data corresponding to the spoken registration name with the audio data processor;

store the identified second utterance in the memory;

verify, with a speaker verification module, that the person speaking the registration name is a user registered in a registration database in association with the registration name in response the first and second utterance data stored in the memory corresponding to a predetermined model of a voice of the user registered in the registration database in association with the registration name; and

generate an output with the user interface device to offer services to the person in response to the speaker verification module verifying that the person speaking the registration name is the user.

12. The telemedical device of claim 11, the controller being further configured to:

generate an output with the user interface device to prompt the person to speak a predetermined phrase in response to the speaker verification module identifying that the first and second utterance data in the memory are insufficient to verify the person with the predetermined model of the voice of the user;

generate audio data corresponding to the spoken predetermined phrase with the audio input device;

identify third utterance data in the audio data corresponding to the spoken predetermined phrase with the audio data processor;

store the third utterance data in the memory; and

verify, with the speaker verification module, that the person speaking the registration name is the user registered in the registration database in response to the first, second, and third utterance data stored in the memory corresponding to a predetermined model of a voice of the user registered in the registration database.

13. The telemedical device of claim 12, the audio data processor being further configured to:

store the third utterance data in the memory in response to the third utterance data corresponding to predetermined utterance data of the predetermined phrase.

14. The telemedical device of claim 12, the controller being further configured to:

generate the output with the user interface device to prompt the person to speak the predetermined phrase a second time in response to the third utterance data not corresponding to predetermined utterance data of the predetermined phrase.

15. The telemedical device of claim 12, the user interface device being further configured to:

generate a prompt for the person to speak a health tip as the predetermined phrase.

16. The telemedical device of claim 15, the user interface device being further configured to:

generate a prompt for the person to speak a health tip that is associated with the registration name in the registration database.

17. The telemedical device of claim 12, the controller being further configured to:

verify, with the speaker verification module, that the person speaking the registration name is not the user registered in the registration database in association with the registration name in response to the first, second, and third utterance data stored in the memory not corresponding to the predetermined model of the voice of the user registered in the registration database; and

generate an output with the user interface device to deny services to the person in response to the speaker verification module verifying that the person speaking the registration name is not the user registered in the registration database.

18. The telemedical device of claim 12, the controller being further configured to:

continue to generate an output with the user interface device to prompt the person to speak at least one additional predetermined phrase in response to the speaker verification module identifying that the first, second, and third utterance data stored in the memory are insufficient to verify the first, second, and third utterance data stored in the memory correspond with the predetermined model of the voice of the user registered in the registration database;

generate audio data corresponding to the at least one additional predetermined phrase spoken by the person with the audio input device;

identify at least one additional utterance data in the audio data corresponding to the at least one additional predetermined phrase with the audio data processor;

store the at least one additional utterance data in the memory; and

verify, with the speaker verification module, that the person is the user registered in the registration database in association with the registration name in response the first, second, third, and at least one additional utterance data stored in the memory corresponding to the predetermined model of the voice of the user registered in the registration database.

19. The telemedical device of claim 18, the controller being further configured to:

identify that the speaker verification module is unable to verify that the person is the user associated with the registration name in the registration database in response to the memory having insufficient utterance data for the speaker verification module to verify that the person is the user after a number of additional utterance data exceeding a predetermined threshold are stored in the memory.

20. The telemedical device of claim 19, the controller being further configured to:

prompt the person, with the user interface device, to input information for verification of the person with a user input device that is different from the audio input device in response to the speaker verification module verifying that the person is not the user associated with the registration name in the registration database.

21. The telemedical device of claim 11 further comprising:

a handheld housing in which the audio input device, audio data processor, memory, registration database, speaker verification module, user interface device, and controller are positioned.