METHOD AND DEVICE FOR IMPLEMENTING VOICE INPUT

• US 20150302852A1
• Filed: 12/17/2013
• Published: 10/22/2015
• Est. Priority Date: 12/31/2012
• Status: Active Grant

• Current Assignee: Baidu Online Network Technology (Beijing) Co., Ltd (Baidu Incorporated)
• Original Assignee: Baidu Online Network Technology (Beijing) Co., Ltd (Baidu Incorporated)
• Inventors: Lu, Yangyang, Jia, Lei
• Granted Patent: US 10,199,036 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G10L 15/04 Segmentation; Word boundary detection
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A network device for implementing voice input comprises an input-obtaining module for obtaining voice input information, a sequence-determining module for determining an input character sequence corresponding to the voice input information based on a voice recognition model, an accuracy-determining module for determining appearance-probability information corresponding to word segments in the input character sequence so as to obtain accuracy information of the word segments, and a transmitting module for transmitting, to a user device, the input character sequence and the accuracy information of the word segments corresponding to the voice input information.

Claims

1-19. -19. (canceled)

20. A method comprising implementing voice input in a network device, wherein implementing voice input comprises obtaining voice input information, determining an input character sequence corresponding to the voice input information based on a voice recognition model, determining appearance-probability information corresponding to word segments in the input character sequence so as to obtain accuracy information of the word segments, and transmitting the input character sequence and the accuracy information of the word segments to a user device corresponding to the voice input information.

21. The method of claim 20, further comprising obtaining an access request for alternative items of at least one of the word segments transmitted by the user device, determining one or more alternative items corresponding to at least one of the word segments based on the access request, and transmitting the one or more alternative items to the user device.

22. The method of claim 21, wherein determining one or more alternative items corresponding to at least one of the word segments based on the access request comprises determining, based on the access request in conjunction with context information of at least one of the word segments, one or more alternative items corresponding to at least one of the word segments.

23. The method of claim 20, wherein determining appearance probability information corresponding to word segments in the input character sequence so as to obtain accuracy information of the word segments comprises determining a conditional probability of the word segments in the input character sequence as appearance probability information of the word segments, and determining accuracy information of the word segments based on an appearance probability threshold value and the appearance probability information of the word segments.

24. The method of claim 23, further comprising determining the appearance-probability threshold value based on the appearance-probability information of the word segments and the appearance probability information of at least one candidate word segment corresponding to the word segments.

25. The method of claim 20, wherein determining an input character sequence corresponding to the voice input information based on a voice recognition model comprises determining an input character sequence corresponding to the voice input information based on a voice recognition model in conjunction with context information corresponding to the voice input information.

26. A method comprising implementing voice input subsidiarily in a user device, wherein implementing voice input subsidiarily in a user device comprises obtaining an input character sequence corresponding to voice input information transmitted by a network device and accuracy information of word segments in the input character sequence, and providing the input character sequence to a user based on accuracy information of the word segments.

27. The method of claim 26, further comprising obtaining the user'"'"'s request operation on alternative items of at least one word segment in the input character sequence, transmitting an access request for alternative items of at least one of the word segments to the network device based on the request operation, receiving one or more alternative items transmitted by the network device based on the access request, and providing at least one of the one or more alternative items to the user.

28. The method of claim 27, further comprising obtaining the user'"'"'s select operation on at least one of the one or more alternative items, and replacing the corresponding word segments in the input character sequence based on the alternative items corresponding to the select operation so as to obtain the updated input character sequence.

29. An apparatus comprising a network device for implementing voice input, wherein the network device comprises an input-obtaining module for obtaining voice input information, a sequence-determining module for determining an input character sequence corresponding to the voice input information based on a voice recognition model, an accuracy-determining module for determining appearance-probability information corresponding to word segments in the input character sequence so as to obtain accuracy information of the word segments, and a transmitting module for transmitting, to a user device, the input character sequence and the accuracy information of the word segments corresponding to the voice input information.

30. The apparatus of claim 29, wherein the network device further comprises a request-obtaining module for obtaining an access request for alternative items of at least one of the word segments transmitted by the user device, an alternative-determining module for determining one or more alternative items corresponding to at least one of the word segments based on the access request, and an alternative transmitting-module for transmitting the one or more alternative items to the user device.

31. The apparatus of claim 30, wherein the alternative determining module is configured for determining, based on the access request in conjunction with context information of at least one of the word segments, one or more alternative items corresponding to at least one of the word segments.

32. The apparatus of claim 29, wherein the accuracy-determining module is configured for determining a conditional probability of the word segments in the input character sequence as appearance-probability information of the word segments, and determining accuracy information of the word segments based on an appearance-probability threshold value and the appearance probability information of the word segments.

33. The apparatus of claim 32, wherein the device further comprises a threshold-value determining module for determining the appearance probability threshold value based on the appearance probability information of the word segments and the appearance probability information of at least one candidate word segment corresponding to the word segments.

34. The apparatus of claim 29, wherein the sequence-determining module is configured for determining an input character sequence corresponding to the voice input information based on a voice recognition model in conjunction with context information corresponding to the voice input information.

35. An apparatus comprising a user device for implementing voice input subsidiarily, wherein the user device comprises a sequence obtaining module for obtaining an input character sequence corresponding to voice input information transmitted by a network device and accuracy information of word segments in the input character sequence, and a providing module for providing the input character sequence to a user based on accuracy information of the word segments.

36. The apparatus of claim 35, wherein the user device further comprises an alternative request obtaining module for obtaining the user'"'"'s request operation on alternative items of at least one word segment in the input character sequence, an access request transmitting module for transmitting an access request for alternative items of at least one of the word segments to the network device based on the request operation, an alternative receiving module for receiving one or more alternative items transmitted by the network device based on the access request, and an alternative providing module for providing at least one of the one or more alternative items to the user.

37. The apparatus of claim 36, wherein the user device further comprises an operation-obtaining module for obtaining the user'"'"'s select operation on at least one of the one or more alternative items, and a replacing module for replacing the corresponding word segments in the input character sequence based on the alternative items corresponding to the select operation so as to obtain the updated input character sequence.