최소 단어 이상 선택하여야 합니다.
최대 10 단어까지만 선택 가능합니다.
다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
NTIS 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
DataON 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Edison 바로가기다음과 같은 기능을 한번의 로그인으로 사용 할 수 있습니다.
Kafe 바로가기국가/구분 | United States(US) Patent 등록 |
---|---|
국제특허분류(IPC7판) |
|
출원번호 | US-0139283 (2009-12-10) |
등록번호 | US-9959870 (2018-05-01) |
국제출원번호 | PCT/GB2009/051684 (2009-12-10) |
§371/§102 date | 20110826 (20110826) |
국제공개번호 | WO2010/067118 (2010-06-17) |
발명자 / 주소 |
|
출원인 / 주소 |
|
대리인 / 주소 |
|
인용정보 | 피인용 횟수 : 1 인용 특허 : 1649 |
A system and method of speech recognition involving a mobile device. Speech input is received (202) on a mobile device (102) and converted (204) to a set of phonetic symbols. Data relating to the phonetic symbols is transferred (206) from the mobile device over a communications network (104) to a re
A system and method of speech recognition involving a mobile device. Speech input is received (202) on a mobile device (102) and converted (204) to a set of phonetic symbols. Data relating to the phonetic symbols is transferred (206) from the mobile device over a communications network (104) to a remote processing device (106) where it is used (208) to identify at least one matching data item from a set of data items (114). Data relating to the at least one matching data item is transferred (210) from the remote processing device to the mobile device and presented (214) thereon.
1. A method of speech recognition involving a mobile device, the method including: receiving speech input on a mobile device;converting the speech input into a set of phonetic symbols on the mobile device;transferring data relating to the phonetic symbols from the mobile device over a communications
1. A method of speech recognition involving a mobile device, the method including: receiving speech input on a mobile device;converting the speech input into a set of phonetic symbols on the mobile device;transferring data relating to the phonetic symbols from the mobile device over a communications network to a remote processing device;using the data relating to the phonetic symbols on the remote processing device to identify at least one matching data item from a set of data items;transferring data relating to the at least one matching data item from the remote processing device to the mobile device; andpresenting data relating to the at least one transferred matching data item using the mobile device. 2. A method according to claim 1, wherein the step of identifying at least one matching data item includes matching the data relating to the phonetic symbols against a set of phonetic reference forms corresponding to the set of data items. 3. A method according to claim 1, wherein the set of phonetic symbols comprises at least one sequence (or lattice) of phonetic symbols. 4. A method according to claim 2, wherein a number of the matching data item(s) transferred from the remote processing device to the mobile device corresponds to a lower of: a maximum number of said data items to be presented on the mobile device, or a number of data items arranged in order of decreasing posterior probability of corresponding to the phonetic symbols down to a predetermined threshold value. 5. A method according to claim 4, wherein the posterior probability for a said data item is computed by taking a match score of the phonetic symbols against the phonetic reference form to the data item and normalizing the match score. 6. A method according to claim 1, wherein the step of presenting data relating to the at least one transferred matching data item comprises displaying an orthographic representation of the at least one transferred matching data item. 7. A method according to claim 1, wherein the step of presenting data relating to the at least one transferred matching data item comprises outputting data corresponding to a map coordinate of a particular location represented by a said transferred matching data item. 8. A method according to claim 1, wherein the step of presenting data relating to the at least one transferred matching data item comprises outputting data corresponding to an identification code for a piece of music or media represented by the transferred matching data item. 9. A method according to claim 1, wherein the step of presenting data relating to the at least one transferred matching data item comprises using a speech synthesis technique executing on the mobile device to generate a spoken form of the at least one transferred matching data item. 10. A method according to claim 1, further including the mobile device storing data representing the speech input and the mobile device performing a rescoring process using the at least one transferred matching data item and the stored speech input data. 11. A method according to claim 10, wherein the rescoring process includes (for each said transferred matching data item to be rescored) generating sequences or lattices of acoustic hidden Markov models (HMMs) of phonetic units corresponding to a phonetic specification of reference pronunciations corresponding to a said transferred matching data item. 12. A method according to claim 11, wherein each of the sequences or lattices is matched against a sequence of frames of spectrum parameters corresponding to the speech input data stored in the mobile device to produce a match score. 13. A method according to claim 12, wherein the matching of the sequences or lattices involves Viterbi time alignment or a full forward probability method. 14. A method according claim 10, wherein the rescoring process involves producing a network including data representing phonetic specifications of all of the data items in the set. 15. A method according to claim 11, wherein the rescoring process includes the mobile device sharing common elements in the phonetic specification in order to reduce an amount of data transferred to the mobile device for the rescoring process. 16. A method according to claim 11, wherein the rescoring process includes: transferring data specifying a general-purpose sub-grammar representing a set of alternative number-word sequences to the mobile device for the rescoring process, andthe mobile device using the sub-grammar to determine a most likely one of the number-words sequences to be presented. 17. A method according to claim 11, wherein the rescoring process includes: transferring phonetic specifications corresponding to the matching data items from the remote processing device to the mobile device with corresponding indices instead of a full description of the matching data items;the mobile device selecting which of the phonetic specification(s) is/are to be presented;transferring the index of indices corresponding to the selected phonetic specifications(s) from the mobile device to the remote processing device;the remote processing device transferring data representing a full description(s) of the data item(s) corresponding to the index or indices transferred by the mobile device back to the mobile device; andthe mobile device presenting the data representing the full description(s). 18. A method according to claim 11, wherein the rescoring process includes: transferring from the remote processing device the matching data items corresponding to a predetermined number of best matches of the phonetic symbols and the data items in the set. 19. One or more non-transitory computer-readable storage media storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of a mobile device, causes the mobile device to: receive speech input on the mobile device;convert the speech input into a set of phonetic symbols on the mobile device;transfer data relating to the phonetic symbols from the mobile device over a communications network to a remote processing device;use the data relating to the phonetic symbols on the remote processing device to identify at least one matching data item from a set of data items;transfer data relating to the at least one matching data item from the remote processing device to the mobile device; andpresent data relating to the at least one transferred matching data item using the mobile device. 20. A system adapted to perform speech recognition involving a mobile device, the system including: a mobile device configured to receive speech input;the mobile device including a device configured to convert the speech input to a set of phonetic symbols on the mobile device;the mobile device further configured to transfer data relating to the phonetic symbols to a remote processing device over a communications network;the remote processing device including a device configured to use the data relating to the phonetic symbols to identify at least one matching data item from a set of data items;the remote device including a communications device configured to transfer data relating to the at least one matching data item to the mobile device over the communications network; andthe mobile device including a device configured to present data relating to the at least one transferred matching data item. 21. A system according to claim 20, wherein the mobile device comprises a mobile telephone. 22. A system according to claim 20, wherein the communications network comprises a GSM, CDMA or UMTS network. 23. A system according to claim 20, wherein the remote processing device further includes a device configured to match the data relating to the phonetic symbols against a set of phonetic reference forms corresponding to the set of data items. 24. A system according to claim 20, wherein the set of phonetic symbols comprises at least one sequence (or lattice) of phonetic symbols. 25. A system according to claim 23, wherein a number of the matching data item(s) transferred from the remote processing device to the mobile device corresponds to a lower of: a maximum number of said data items to be presented on the mobile device, or a number of data items arranged in order of decreasing posterior probability of corresponding to the phonetic symbols down to a predetermined threshold value. 26. A system according to claim 25, wherein the posterior probability for a said data item is computed by taking a match score of the phonetic symbols against the phonetic reference form to the data item and normalizing the match score. 27. A system according to claim 20, wherein the mobile device further includes a device configured to display an orthographic representation of the at least one transferred matching data item. 28. A system according to claim 20, wherein the mobile device further includes a device configured to output data corresponding to a map coordinate of a particular location represented by a said transferred matching data item. 29. A system according to claim 20, wherein the mobile device further includes a device configured to output data corresponding to an identification code for a piece of music or media represented by the transferred matching data item. 30. A system according to claim 20, wherein the mobile device further includes a device configured to use a speech synthesis technique to generate a spoken form of the at least one transferred matching data item. 31. A system according to claim 20, wherein the mobile device is further configured to store data representing the speech input and perform a rescoring process using the at least one transferred matching data item and the stored speech input data. 32. A system according to claim 31, wherein the rescoring process includes (for each said transferred matching data item to be rescored) generating sequences or lattices of acoustic hidden Markov models (HMMs) of phonetic units corresponding to a phonetic specification of reference pronunciations corresponding to a said transferred matching data item. 33. A system according to claim 32, wherein each of the sequences or lattices is matched against a sequence of frames of spectrum parameters corresponding to the speech input data stored in the mobile device to produce a match score. 34. A system according to claim 33, wherein the matching of the sequences or lattices involves Viterbi time alignment or a full forward probability method. 35. A system according claim 31, wherein the rescoring process involves producing a network including data representing phonetic specifications of all of the data items in the set. 36. A system according to claim 32, wherein the rescoring process includes the mobile device sharing common elements in the phonetic specification in order to reduce an amount of data transferred to the mobile device for the rescoring process. 37. A system according to claim 32, wherein the rescoring process includes: transferring data specifying a general-purpose sub-grammar representing a set of alternative number-word sequences to the mobile device for the rescoring process, andusing, by the mobile device, the sub-grammar to determine a most likely one of the number-words sequences to be presented. 38. A system according to claim 32, wherein the rescoring process includes: transferring phonetic specifications corresponding to the matching data items from the remote processing device to the mobile device with corresponding indices instead of a full description of the matching data items;selecting, by the mobile device, which of the phonetic specification(s) is/are to be presented;transferring the index of indices corresponding to the selected phonetic specifications(s) from the mobile device to the remote processing device;transferring, by the remote processing device, data representing a full description(s) of the data item(s) corresponding to the index or indices transferred by the mobile device back to the mobile device; andpresenting, by the mobile device, the data representing the full description(s). 39. A system according to claim 32, wherein the rescoring process includes: transferring, from the remote processing device, the matching data items corresponding to a predetermined number of best matches of the phonetic symbols and the data items in the set. 40. The one or more non-transitory computer-readable storage media of claim 19, wherein identifying at least one matching data item includes matching the data relating to the phonetic symbols against a set of phonetic reference forms corresponding to the set of data items. 41. The one or more non-transitory computer-readable storage media of claim 19, wherein the set of phonetic symbols comprises at least one sequence (or lattice) of phonetic symbols. 42. The one or more non-transitory computer-readable storage media of claim 40, wherein a number of the matching data item(s) transferred from the remote processing device to the mobile device corresponds to a lower of: a maximum number of said data items to be presented on the mobile device, or a number of data items arranged in order of decreasing posterior probability of corresponding to the phonetic symbols down to a predetermined threshold value. 43. The one or more non-transitory computer-readable storage media of claim 42, wherein the one or more programs further comprise instructions, which when executed by the one or more processors of the mobile device, causes the mobile device to compute the posterior probability for a said data item by taking a match score of the phonetic symbols against the phonetic reference form to the data item and normalizing the match score. 44. The one or more non-transitory computer-readable storage media of claim 19, wherein the one or more programs further comprise instructions, which when executed by the one or more processors of the mobile device, causes the mobile device to display, during the step of presenting data relating to the at least one transferred matching data item, an orthographic representation of the at least one transferred matching data item. 45. The one or more non-transitory computer-readable storage media of claim 19, wherein the one or more programs further comprise instructions, which when executed by the one or more processors of the mobile device, causes the mobile device to output, during the step of presenting data relating to the at least one transferred matching data item, data corresponding to a map coordinate of a particular location represented by a said transferred matching data item. 46. The one or more non-transitory computer-readable storage media of claim 19, wherein the one or more programs further comprise instructions, which when executed by the one or more processors of the mobile device, causes the mobile device to output, during the step of presenting data relating to the at least one transferred matching data item, data corresponding to an identification code for a piece of music or media represented by the transferred matching data item. 47. The one or more non-transitory computer-readable storage media of claim 19, wherein the one or more programs further comprise instructions, which when executed by the one or more processors of the mobile device, causes the mobile device to use, during the step of presenting data relating to the at least one transferred matching data item, a speech synthesis technique executing on the mobile device to generate a spoken form of the at least one transferred matching data item. 48. The one or more non-transitory computer-readable storage media of claim 19, wherein the one or more programs further comprise instructions, which when executed by the one or more processors of the mobile device, causes the mobile device to store data representing the speech input and perform a rescoring process using the at least one transferred matching data item and the stored speech input data. 49. The one or more non-transitory computer-readable storage media of claim 48, wherein the one or more programs further comprise instructions, which when executed by the one or more processors of the mobile device, causes the mobile device to generate, during the rescoring process (for each said transferred matching data item to be rescored), sequences or lattices of acoustic hidden Markov models (HMMs) of phonetic units corresponding to a phonetic specification of reference pronunciations corresponding to a said transferred matching data item. 50. The one or more non-transitory computer-readable storage media of claim 49, wherein the one or more programs further comprise instructions, which when executed by the one or more processors of the mobile device, causes the mobile device to match each of the sequences or lattices against a sequence of frames of spectrum parameters corresponding to the speech input data stored in the mobile device to produce a match score. 51. The one or more non-transitory computer-readable storage media of claim 50, wherein the matching of the sequences or lattices involves Viterbi time alignment or a full forward probability method. 52. The one or more non-transitory computer-readable storage media of claim 48, wherein the rescoring process includes producing a network including data representing phonetic specifications of all of the data items in the set. 53. The one or more non-transitory computer-readable storage media of claim 49, wherein the rescoring process includes sharing, by the mobile device, common elements in the phonetic specification in order to reduce an amount of data transferred to the mobile device for the rescoring process. 54. The one or more non-transitory computer-readable storage media of claim 49, wherein the rescoring process includes: transferring data specifying a general-purpose sub-grammar representing a set of alternative number-word sequences to the mobile device for the rescoring process, andusing, by the mobile device, the sub-grammar to determine a most likely one of the number-words sequences to be presented. 55. The one or ore non-transitory computer-readable storage media of claim 49, wherein the rescoring process includes: transferring phonetic specifications corresponding to the matching data items from the remote processing device to the mobile device with corresponding indices instead of a full description of the matching data items;selecting, by the mobile device, which of the phonetic specification(s) is/are to be presented;transferring the index of indices corresponding to the selected phonetic specifications(s) from the mobile device to the remote processing device;transferring, by the remote processing device, data representing a full description(s) of the data item(s) corresponding to the index or indices transferred by the mobile device back to the mobile device; andpresenting, by the mobile device, the data representing the full description(s). 56. The one or more non-transitory computer-readable storage media of claim 49, wherein the rescoring process includes transferring, from the remote processing device, the matching data items corresponding to a predetermined number of best matches of the phonetic symbols and the data items in the set.
Copyright KISTI. All Rights Reserved.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.