System, method and device for language education through a voice portal

• US 8,371,857 B2
• Filed: 05/02/2012
• Issued: 02/12/2013
• Est. Priority Date: 05/29/2003
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Raya, Madhuri, Funk, Karsten, Ravula, Sharmila, Meng, Yao
• Primary Examiner(s): Thai, Xuan
• Assistant Examiner(s): Hadizonooz, Banafsheh
• Application Number:
  US13/462,536
• Publication Number:
  US 20120214141A1
• Time in Patent Office: 286 Days
• Field of Search: 434/307, 434/156, 434/157, 434/319, 434/318, 283/46, 704/200
• US Class Current: 434/156
• CPC Class Codes: G09B 19/04 Speaking with audible presentation of the material to be studied G09B5/04
• CPC Class Codes: G09B 5/06 with both visual and audible presentation of the material to be studied

Abstract

A method of teaching pronunciation is provided which includes communicating by a voice portal server to a user a model word and detecting a response by the user to the voice portal server. The method also includes comparing the response word to the model word and determining a confidence level based on the comparison of the response word to the model word. The method further includes comparing an acceptance limit to the confidence level and confirming a correct pronunciation of the model word if the confidence level one of equals and exceeds the acceptance limit.

Claims

1. A system for teaching a user pronunciation, comprising:

a voice portal arrangement;

a communication arrangement adapted to be coupled with the voice portal arrangement;

an application server adapted to be coupled with the voice portal arrangement; and

a geo-server for determining a location;

wherein;

the geo-server determines a location of the user and a dominant language and a regional dialect prevailing at the determined location of the user;

the voice portal arrangement provides the user a prompt in the determined regional dialect of the dominant language, and compares at least one word spoken by the user in response to the prompt to determine a confidence lever;

the at least one word spoken by the user is associated in memory with a plurality of valid pronunciations; and

one of the plurality of valid pronunciations used by the user for the at least one word spoken by the user is determined to be incorrect based on the regional dialect.

2. The system of claim 1, wherein the communication arrangement is adapted to be wirelessly coupled with the voice portal server.

3. The system of claim 1, wherein the communication arrangement is adapted to be coupled with the voice portal server.

4. The system of claim 1, wherein the communication arrangement is one of a telephone, a mobile phone, a wireless communication arrangement and an internet-enabled arrangement.

5. The system of claim 1, wherein the application server at least one of includes a language database and communicates with the language database, the language database providing the phrase.

6. The system of claim 1, wherein the application server stores a user profile, the user profile including at least one of an instructional history and a proficiency level.

7. The system of claim 1, wherein a telephone call, in which the voice portal arrangement receives the at least one spoken word, is established by the user dialing a telephone number to access the voice portal arrangement and the voice portal arrangement answering the telephone call.

8. The system of claim 1, wherein a user'"'"'s language proficiency is assessed based at least in part on the determined correctness of the pronunciation, and a lesson plan is selected for the user based on the assessed language proficiency of the user.

9. The system of claim 1, wherein the location of the user is automatically determined via one of a GPS system and a mobile phone location system.

10. The system of claim 1, wherein a geo-server identifies one of the dominant language and the regional dialect based on the location of the user.

11. The system of claim 1, wherein the correctness of the spoken word is determined based on the regional dialect.

12. The system of claim 1, wherein the location of the user is a geographic location.

13. The system of claim 1, wherein the system is configured to output to the user data for correcting the user'"'"'s pronunciation where the pronunciation used by the user is determined to be incorrect based on the one of the dominant language and regional dialect.

14. A method of teaching a user a language, comprising:

determining a location of the user and a dominant language and a regional dialect prevailing at the determined location of the user;

communicating a prompt in the determined regional dialect of the dominant language to a user by a voice portal;

detecting a response provided by the user, the response being provided to the voice portal;

parsing the response into at least one uttered phrase, each of the at least one uttered phrase being associated with at least one corresponding slot;

comparing each of the at least one uttered phrase with at least one stored phrase, the at least one stored phrase being associated with the at least one corresponding slot;

determining a confidence level by comparing each of the at least one uttered phrase with each of the at least one stored phrase;

comparing each determined confidence level to an associated acceptance limit, the associated acceptance limit being associated with each of the at least one stored phrase; and

confirming that each of the at least one uttered phrase corresponds to the at least one stored phrase if the confidence level is not less than the associated acceptance limit;

wherein a pronunciation of the at least one uttered phrase is determined (a) to be one of a stored plurality of valid pronunciations, and (b) to be incorrect based on the regional dialect.

15. The method of claim 14, further comprising:

informing the user by the voice portal server of an incorrect usage if each determined confidence level is less than each associated acceptance limit for any slot.

16. The method of claim 15, further comprising:

repeating the prompt to the user by the voice portal server.

17. The method of claim 15, further comprising:

informing the user by the voice portal server of the at least one uttered phrase associated with the at least one corresponding slot for which each determined confidence level is less than each associated acceptance limit.

18. The method of claim 14, further comprising:

informing the user of a correct usage if any confidence level is not less than the associated acceptance limit for all slots.

19. The method of claim 14, wherein the comparing of each of the at least one uttered phrase with the at least one stored phrase includes accessing the at least one stored phrase from a database.

20. The method of claim 14, wherein the parsing of the response into the at least one uttered phrase is performed using a recognized silence as a separator between two adjacent uttered phrases.

21. The method of claim 14, wherein the at least one stored phrase includes at least one incorrectly pronounced phrase.

22. The method of claim 21, further comprising:

informing the user of an incorrect pronunciation if the confidence level is not less than the acceptance limit for the at least one incorrectly pronounced phrase.

23. The method of claim 22, wherein the informing of the user of the incorrect pronunciation includes informing the user of at least one incorrectly pronounced uttered phrase, the at least one incorrectly pronounced uttered phrase corresponding to all uttered phrases with corresponding confidence levels that are not less than the associated acceptance limit for the at least one incorrectly pronounced phrase.

24. The method of claim 14, wherein the prompt includes one of a scenario and at least one word.

25. A non-transitive computer-readable medium, having a program executable on a processor arrangement, to perform a method, the method comprising:

determining a location of the user and a dominant language and a regional dialect prevailing at the determined location of the user;

communicating a prompt in the determined regional dialect of the dominant language to a user by a voice portal;

detecting a response provided by the user, the response being provided to the voice portal;

parsing the response into at least one uttered phrase, each of the at least one uttered phrase being associated with at least one corresponding slot;

comparing each of the at least one uttered phrase with at least one stored phrase, the at least one stored phrase being associated with the at least one corresponding slot;

determining a confidence level by comparing each of the at least one uttered phrase with each of the at least one stored phrase;

comparing each determined confidence level to an associated acceptance limit, the associated acceptance limit being associated with each of the at least one stored phrase; and

confirming that each of the at least one uttered phrase corresponds to the at least one stored phrase if the confidence level is not less than the associated acceptance limit;

wherein a pronunciation of the at least one uttered phrase is determined (a) to be one of a stored plurality of valid pronunciations, and (b) to be incorrect based on the regional dialect.