Location-based speech recognition for preparation of electronic tax return

• US 9,946,699 B1
• Filed: 08/29/2012
• Issued: 04/17/2018
• Est. Priority Date: 08/29/2012
• Status: Active Grant

• Current Assignee: Intuit, Inc.
• Original Assignee: Intuit, Inc.
• Inventors: Dye, Christopher M., Zuberi, Azhar M., McVickar, Richard E.
• Primary Examiner(s): Jackson, Jakieda
• Application Number:
  US13/598,546
• Time in Patent Office: 2,057 Days
• Field of Search: 704235
• CPC Class Codes: G06F 40/174 Form filling; Merging
• CPC Class Codes: G06Q 30/04 Billing or invoicing
• CPC Class Codes: G06Q 40/123 Tax preparation or submission
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

Methods, systems and articles of manufacture for location-based speech recognition for preparation of an electronic tax return.

Claims

1. A computer-implemented method, comprising:

a computing device, by a computerized tax return preparation application comprising computer-executable instructions stored in a data store of the computing device and executed by the processor of the computing device,

receiving respective voice samples of respective users of the tax preparation application;

parsing the voice samples into voice sample terms;

generating a voice sample database comprising;

voice sample terms,

location data associated with respective electronic voice sample terms, and

identification of respective interview screens or tax forms of the computerized tax return preparation application is associated with respective voice sample terms and respective location data;

after generation of the voice sample database, the computing device, by the computerized tax return preparation application;

receiving, from a computing apparatus utilized to prepare an electronic tax return utilizing the computerized tax return preparation application, an electronic output generated by a microphone of the computing apparatus based on voice input comprising the preparer of the electronic tax return speaking into the microphone of the computing apparatus during preparation of the electronic tax return;

determining location data of the electronic output;

parsing the electronic output into voice data terms;

accessing the previously generated voice sample database;

determining a meaning or intent of the voice input of the preparer based at least in part upon voice data terms and associated location data of the received voice data and respective stored voice sample terms and associated location data of the generated voice sample database;

selecting an interview screen or tax form of the computerized tax return preparation application based at least in part upon the determined meaning or intent of the voice input;

automatically populating the selected interview screen or tax form with associated electronic tax return data determined based at least in part upon the electronic output;

executing machine learning by communicating a request for feedback whether the determined meaning or intent of the voice input was what the prepare intended to be conveyed by the voice data, receiving a response from the preparer in response to the request for feedback, and updating the voice sample database based on the received response to improve accuracy of future voice data interpretation during preparation of future electronic tax returns, the response being associated with a previously recorded voice sample and the location data of the associated voice sample;

determining at least one additional interview screen or tax form to present to the user based at least in part upon the generated voice sample database;

presenting the at least one additional interview screen or tax form to the preparer through a display of the computing device without additional preparer input thereby improving how interview screens or tax forms are presented to the preparer by the computerized tax return preparation application during preparation of an the electronic tax return.

2. The method of claim 1, the voice input comprising live voice data spoken by the preparer into the microphone during preparation of the electronic tax return.

3. The method of claim 1, wherein interpretation of the received voice data is based at least in part upon the location data associated with the received voice data of the preparer comprising location data that has been already entered into the electronic tax return being prepared utilizing the computing apparatus.

4. The method of claim 3, the location data that has been already entered into the electronic tax return being prepared being selected from at least one of the group consisting of a city, a state, a zip code and a country.

5. The method of claim 1, the location data of the received voice input being determined by a computerized voice analyzer element of the computerized tax return preparation application independently of the electronic tax return being prepared, wherein the location data of the received voice input is based at least in part upon location data of the computing apparatus utilized to prepare the electronic tax return.

6. The method of claim 5, the location data of the computing apparatus utilized to prepare the electronic tax return comprising global positioning system location data of the computing apparatus.

7. The method of claim 5, the location data of the computing apparatus utilized to prepare the electronic tax return comprising an internet protocol address assigned to the computing apparatus.

8. The method of claim 1, the location data of the received voice input being determined by a computerized voice analyzer element of the computerized tax return preparation application and independently of electronic data of the electronic tax return.

9. The method of claim 1, a computerized voice analyzer element of the computerized tax return preparation application interpreting the received voice data comprising:

selecting a previously recorded voice sample of another person in the data store as a potential match to the received voice input based at least in part upon the location data of the received voice input and location data of the selected voice sample of another person;

determining a meaning of the selected voice sample based least in part upon the location data of the received voice data and voice sample location data; and

interpreting the received voice data based at least in part upon the determined meaning.

10. The method of claim 9, the voice samples and respective location data of respective voice samples stored in the data store reflecting how different persons at different locations speak the same terms related to the electronic tax return differently, wherein similarities between speech patterns in the received voice data and previously recorded voice samples stored in the data store are determined by the computerized voice analyzer element to interpret the received voice data.

11. The method of claim 9, selecting the previously recorded voice sample comprising the computerized voice analyzer element:

filtering previously recorded voice samples based at least in part upon the location data of the received voice data of the preparer to select previously recorded voice samples;

comparing the received voice data and the selected previously recorded voice samples; and

selecting a voice sample based at least in part upon the comparison,

wherein interpreting the received voice data is based at least in part upon the selected voice sample.

12. The method of claim 9, wherein multiple previously recorded voice samples stored in the data store are selected by the computerized voice analyzer element as potential matches, the method further comprising the computerized voice analyzer element:

receiving an indication of a stored voice sample selected by the preparer,

wherein the meaning of the preparer selected voice sample is determined, and the interpretation of the received voice data is based at least in part upon the determined meaning of the preparer selected voice sample.

13. The method of claim 9, further comprising the computerized voice analyzer element receiving and storing voice samples of respective persons and associated with a term utilized to prepare the electronic tax return term, the respective voice samples being previously spoken by different other persons in respective different geographic locations.

14. The method of claim 13, wherein the different other persons are other users of the computerized tax return preparation application.

15. The method of claim 1, the response indicating that the interpretation was correct, the method further comprising the computerized tax return preparation application updating the voice sample database by storing the received voice data to the voice sample database, the received voice data being associated with the meaning of the identified voice sample.

16. The method of claim 15, further comprising the computerized tax return preparation application updating the voice sample database by adding the location data of the of the received voice data to the location data of the identified previously recorded voice sample and the meaning of the identified previously recorded voice sample.

17. The method of claim 1, the response indicating that a first interpretation was incorrect and providing a second interpretation relative to the electronic tax return, the method further comprising the computerized tax return preparation application updating the voice sample database by:

identifying another previously recorded voice sample stored in the voice sample database in the data store having another meaning corresponding to the second interpretation; and

storing the received voice data to the voice sample database in the data store, the received voice data being associated with the other identified previously recorded voice sample and the other meaning.

18. The method of claim 17, further comprising the computerized tax return preparation application updating the voice sample database by adding the location data of the received voice data to location data of the other identified previously recorded voice sample and the other meaning.

19. The method of claim 17, further comprising the computerized tax return preparation application updating the voice sample database by deleting location data that was initially associated with the first interpretation based at least in part upon the first interpretation of the voice data being incorrect.

20. The method of claim 1, further comprising the computerized tax return preparation application:

selecting a first field or section of the electronic tax return associated with the interpretation of the received voice data;

converting the received voice data into tax return data; and

populating the first field or section with the tax return data.

21. The method of claim 20, further comprising, before the first field or section is populated, the computerized tax return preparation application receiving confirmation from the preparer that the selected first field or section is to be populated with the tax return data.

22. The method of claim 20, further comprising:

a computerized voice analyzer element of the computerized tax return preparation application selecting at least one additional field or section of the electronic tax return that is associated with the first field or section and converting the voice data into tax return data; and

the computerized tax return preparation application automatically populating the first field or section and the at least one additional field or section with tax return data.

23. The method of claim 22, further comprising, before the first field or section and the at least one additional field or section are populated, the computerized tax return preparation application receiving confirmation from the preparer that the first field or section and the at least one additional field or section are to be populated with the tax return data.

24. The method of claim 1, wherein the interpretation of the received voice input involves a tax document or interview screen generated by the computerized tax return preparation application, the method further comprising computerized tax return preparation application:

determining questions to be presented to the preparer about the tax form or interview screen; and

presenting the questions to the preparer through a display of the computing apparatus without additional voice input by the preparer.

25. The method of claim 1, wherein the interpretation of the received voice data involves a first tax document or interview screen, the method further comprising the computerized tax return preparation application:

selecting a second tax document or interview screen related to the first tax document or interview screen; and

presenting the second tax document or interview screen to the preparer through a display of the computing apparatus without additional voice input by the preparer.

26. The method of claim 1, wherein interpretation of the received voice sample based at least in part upon the location data of the received voice sample accounts for an accent associated with a location from which the voice sample was received.

27. The method of claim 1, wherein interpretation of the received voice sample based at least in part upon the location data of the received voice sample accounts for a colloquial expression spoken in a location from which the voice sample was received and that is related to a term of the electronic tax return.

28. The method of claim 1, wherein interpretation of the location data of the received voice sample based at least in part upon the location data of the received the voice sample accounts for a pronunciation of a term spoken in a location from which the voice sample was received and that is related to the electronic tax return in the location.

29. The method of claim 1, the location data of the received voice data comprising data of a region, a state or a zip code in which the preparer resides.

30. The method of claim 1, the voice input being interpreted by a computerized voice analyzer element of the computerized tax return preparation application based at least in part upon a comparison of the received electronic output and previously recorded voice samples reflecting how different persons at different locations speak the same terms related to the electronic tax return differently, wherein similarities between speech patterns in the received voice data and previously recorded voice samples stored in the data store are determined by the computerized voice analyzer element to interpret the received voice data.

31. The method of claim 1, the second format comprising frequency domain representations of the voice samples and the received voice data.

32. The method of claim 31, the frequency domain representations being generated by application of a Fourier transformation to the voice samples and the received voice data.