METHODS AND SYSTEMS FOR SPEECH ADAPTATION DATA

• US 20130325452A1
• Filed: 09/10/2012
• Published: 12/05/2013
• Est. Priority Date: 05/31/2012
• Status: Active Grant

• Current Assignee: Elwha LLC (Intellectual Ventures LLC)
• Original Assignee: Elwha LLC (Intellectual Ventures LLC)
• Inventors: Levien, Royce A., Lord, Richard T., Lord, Robert W., Malamud, Mark A.
• Granted Patent: US 9,305,565 B2
• Time in Patent Office: Days
• US Class Current: 704/201
• CPC Class Codes: G10L 15/06 Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice G10L15/14 takes precedence
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/065 Adaptation
• CPC Class Codes: G10L 15/07 to the speaker
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence

Abstract

Computationally implemented methods and systems include receiving speech data correlated to one or more words spoken by a particular party, receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party, obtaining target data regarding a target configured to process at least a portion of the received speech data, and determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data. In addition to the foregoing, other aspects are described in the claims, drawings, and text.

Claims

1. A computationally-implemented method, comprising:

receiving speech data correlated to one or more words spoken by a particular party;

receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party;

obtaining target data regarding a target configured to process at least a portion of the received speech data;

determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data; and

transmitting adaptation result data that is based on at least one aspect of the received speech data.

9. The computationally-implemented method of claim 1, wherein said receiving speech data correlated to one or more words spoken by a particular party comprises:

receiving speech data comprising a representation of speech that is spoken by the particular party.

13. The computationally-implemented method of claim 9, wherein said receiving speech data comprising a representation of speech that is spoken by the particular party comprises:

receiving speech data corresponding to received speech spoken by the particular party that has been at least partially processed.

25. The computationally-implemented method of claim 1, wherein said receiving speech data correlated to one or more words spoken by a particular party comprises:

receiving, from a further device, speech data correlated to one or more words spoken by a particular party.

26. The computationally-implemented method of claim 25, wherein said receiving, from a further device, speech data correlated to one or more words spoken by a particular party comprises:

receiving, from the further device, audio data derived from one or more words spoken by the particular party.

29. The computationally-implemented method of claim 26, wherein said receiving, from the further device, audio data derived from one or more words spoken by the particular party comprises:

receiving, from the further device, audio data derived by the further device from one or more words spoken by the particular party and detected by the particular device.

49. The computationally-implemented method of claim 1, wherein said receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party comprises:

receiving adaptation data from a further device, said adaptation data at least partly based on at least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party.

50. The computationally-implemented method of claim 49, wherein said receiving adaptation data from a further device, said adaptation data at least partly based on at least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party comprises:

receiving adaptation data from a further device, said adaptation data originating at the further device and at least partly based on least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party.

51. The computationally-implemented method of claim 49, wherein said receiving adaptation data from a further device, said adaptation data at least partly based on at least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party comprises:

receiving adaptation data from a further device related to the particular device, said adaptation data originating at the further device and at least partly based on least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party.

52. The computationally-implemented method of claim 51, wherein said receiving adaptation data from a further device related to the particular device, said adaptation data originating at the further device and at least partly based on least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party comprises:

receiving adaptation data from a further device associated with the particular party, said adaptation data originating at the further device and at least partly based on least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party.

55. The computationally-implemented method of claim 49, wherein said receiving adaptation data from a further device, said adaptation data at least partly based on at least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party comprises:

receiving adaptation data from a further device, said adaptation data received by the further device from the particular device, and said adaptation data at least partly based on least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on the particular device associated with the particular party.

71. The computationally-implemented method of claim 1, wherein said receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party comprises:

receiving adaptation data that is at least partly based on at least one speech interaction of the particular party that used one or more same utterances as speech used in the detected speech data, said one or more same utterances spoken to a different device than a target device to which the detected speech data is directed.

74. The computationally-implemented method of claim 1, wherein said receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party comprises:

acquiring a sentence diagramming path selection algorithm based on at least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party.

75. The computationally-implemented method of claim 1, wherein said receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party comprises:

receiving adaptation data that is at least partly based on at least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data was collected by the particular device associated with the particular party.

80. The computationally-implemented method of claim 1, wherein said receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party comprises:

receiving adaptation data that is at least partly based on at least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data is temporarily stored on the particular device associated with the particular party until it is deposited at a remote server.

81. The computationally-implemented method of claim 1, wherein said receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party comprises:

receiving adaptation data that is at least partly based on at least one speech interaction of the particular party that is discrete from the detected speech data, wherein at least a portion of the adaptation data was transmitted from a first device to a second device using the particular device associated with the particular party as a conduit configured to facilitate the transmission.

84. The computationally-implemented method of claim 1, wherein said receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party comprises:

receiving adaptation data that is at least partly based on at least one speech interaction of the particular party that is discrete from the detected speech data; and

adding further data to the received adaptation data.

91. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

receiving an indication from a speech processing component that the adaptation data is configured to be applied to the target device to assist in processing at least a portion of the received speech data.

96. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

generating target data regarding a target device configured to process at least a portion of the received speech data.

97. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

determining whether the speech data is configured to be processed by a speech recognition component to which the adaptation data has been applied; and

generating target data regarding the intended target device based on the determination.

98. The computationally-implemented method of claim 97, wherein said generating target data regarding the intended target device based on the determination comprises:

analyzing at least a portion of the speech data; and

determining target data regarding the intended target device at least partly based on a result of analyzing the at least a portion of the received speech data.

99. The computationally-implemented method of claim 98, wherein said analyzing at least a portion of the speech data comprises:

analyzing at least a portion of the speech data using a speech recognition component to which the adaptation data has been applied.

100. The computationally-implemented method of claim 99, wherein said analyzing at least a portion of the speech data using a speech recognition component to which the adaptation data has been applied comprises:

analyzing at least a portion of the speech data using a speech recognition component of the target device to which the adaptation data has been applied.

101. The computationally-implemented method of claim 99, wherein said analyzing at least a portion of the speech data using a speech recognition component to which the adaptation data has been applied comprises:

analyzing at least a portion of the speech data using a speech recognition component of a further device to which the adaptation data has been applied.

104. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

determining whether a target device is configured to process the received speech data; and

generating target data regarding the target device based on the determination regarding the target device.

105. The computationally-implemented method of claim 104, wherein said determining whether a target device is configured to process the received speech data comprises:

analyzing at least a portion of the speech data; and

determining that the target device is configured to process the speech data at least partly based on the analysis of the speech data.

106. The computationally-implemented method of claim 104, wherein said determining whether a target device is configured to process the received speech data comprises:

extracting header data from a header of the received speech data indicating a type of the intended target device that is configured to process the received speech data; and

determining that a type of the target device is a same type as the type of the intended target device.

107. The computationally-implemented method of claim 106, wherein said extracting header data from a header of the received speech data indicating a type of the intended target device that is configured to process the received speech data comprises:

extracting header data from a header of the received speech data indicating a manufacturer of one or more intended target devices that are configured to process the received speech data.

111. The computationally-implemented method of claim 104, wherein said determining whether a target device is configured to process the received speech data comprises:

converting the received speech data into data that is recognizable by a target device.

112. The computationally-implemented method of claim 111, wherein said converting the received speech data into data that is recognizable by a target device comprises:

converting the received speech data into one or more commands or command modifiers configured to be recognized by a control component of the target device.

113. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

receiving target data regarding a target device configured to process at least a portion of the received speech data.

115. The computationally-implemented method of claim 113, wherein said receiving target data regarding a target device configured to process at least a portion of the received speech data comprises:

receiving target data regarding a target device configured to process at least a portion of the received speech data from a further device that is different than the particular device.

121. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

receiving data identifying the target device configured to process at least a portion of the received speech data.

124. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

receiving an address of the target device configured to process at least a portion of the received speech data.

126. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

acquiring target data regarding an intended application module configured to process at least a portion of the received speech data.

127. The computationally-implemented method of claim 126, wherein said acquiring target data regarding an intended application module configured to process at least a portion of the received speech data comprises:

acquiring target data regarding an intended application module configured to process at least a portion of the received speech data, wherein at least a portion of said processing is configured to be facilitated by the adaptation data.

128. The computationally-implemented method of claim 126, wherein said acquiring target data regarding an intended application module configured to process at least a portion of the received speech data comprises:

acquiring target data regarding a speech data processing capability of an intended application module configured to process at least a portion of the received speech data.

129. The computationally-implemented method of claim 1, wherein said obtaining target data regarding a target configured to process at least a portion of the received speech data comprises:

acquiring target data regarding a first application module configured to process at least a portion of the received speech data, and a second application module configured to process at least a portion of the received speech data.

130. The computationally-implemented method of claim 129, wherein said acquiring target data regarding a first application module configured to process at least a portion of the received speech data, and a second application module configured to process at least a portion of the received speech data comprises:

acquiring target data regarding a word processing application module configured to process at least a portion of the received speech data, and a speech recognition application module configured to process at least a portion of the received speech data.

131. The computationally-implemented method of claim 129, wherein said acquiring target data regarding a first application module configured to process at least a portion of the received speech data, and a second application module configured to process at least a portion of the received speech data comprises:

acquiring target data regarding a word processing application module configured to process at least a portion of the received speech data, and an operating system module configured to process at least a portion of the received speech data.

132. The computationally-implemented method of claim 129, wherein said acquiring target data regarding a first application module configured to process at least a portion of the received speech data, and a second application module configured to process at least a portion of the received speech data comprises:

acquiring target data regarding a word processing application module configured to process at least a portion of the received speech data, and a spreadsheet processing application module configured to process at least a portion of the received speech data.

134. The computationally-implemented method of claim 1, wherein said determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data comprises:

determining to apply the adaptation data for processing at least a portion of the received speech data when the acquired target data indicates that the received speech data has arrived at its intended target device.

137. The computationally-implemented method of claim 1, wherein said determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data comprises:

determining against application of the adaptation data for processing at least a portion of the received speech data when the acquired target data indicates that the received speech data has arrived at a device other than its intended target device.

139. The computationally-implemented method of claim 1, wherein said determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data comprises:

determining against application of the adaptation data for processing at least a portion of the received speech data when the acquired target data indicates that there are one or more other devices present that are configured to apply the adaptation data for processing at least a portion of the received speech data.

141. The computationally-implemented method of claim 1, wherein said determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data comprises:

determining whether to apply the adaptation data for processing at least a portion of the received speech data when the acquired target data indicates a presence of one or more applications configured to process the received speech data.

143. The computationally-implemented method of claim 1, wherein said determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data comprises:

determining whether to apply the adaptation data for processing at least a portion of the received speech data based on acquired target data comprising one or more characteristics of one or more applications of the target device.

144. The computationally-implemented method of claim 143, wherein said determining whether to apply the adaptation data for processing at least a portion of the received speech data based on acquired target data comprising one or more characteristics of one or more applications of the target device comprises:

determining whether to apply the adaptation data for processing at least a portion of the received speech data based on acquired target data comprising a detection of one or more applications and corresponding one or more characteristics of the one or more applications.

146. The computationally-implemented method of claim 1, wherein said determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data comprises:

determining whether to apply the adaptation data for processing at least a portion of the received speech data at least partly based on a preference flag set on the one or more applications.

148. The computationally-implemented method of claim 1, wherein said determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data comprises:

determining whether to apply the adaptation data based on a decision by an operating system of a device configured to process at least a portion of the received speech data, when the acquired target data indicates that there are one or more applications present configured to process the received speech data.

156. The computationally-implemented method of claim 1, wherein said transmitting adaptation result data that is based on at least one aspect of the received speech data comprises:

transmitting adaptation result data indicating that a determination was made regarding an intended target of the received speech data.

157. The computationally-implemented method of claim 1, wherein said transmitting adaptation result data that is based on at least one aspect of the received speech data comprises:

transmitting adaptation result data that is based on a measure of success of at least one portion of a speech-facilitated transaction corresponding to the received speech data.

158. The computationally-implemented method of claim 157, wherein said transmitting adaptation result data that is based on a measure of success of at least one portion of a speech-facilitated transaction corresponding to the received speech data comprises:

transmitting adaptation result data that comprises a representation of success of at least one portion of a speech-facilitated transaction corresponding to the received speech data.

160. The computationally-implemented method of claim 158, wherein said transmitting adaptation result data that comprises a representation of success of at least one portion of a speech-facilitated transaction corresponding to the received speech data comprises:

transmitting adaptation result data that comprises a numeric representation of success of at least one portion of a speech-facilitated transaction corresponding to the received speech data.

161. The computationally-implemented method of claim 160, wherein said transmitting adaptation result data that comprises a numeric representation of success of at least one portion of a speech-facilitated transaction corresponding to the received speech data comprises:

transmitting a confidence rate of correct interpretation of at least one portion of the speech-facilitated transaction corresponding to the received speech data.

2. (canceled)

3. (canceled)

4. (canceled)

5. (canceled)

6. (canceled)

7. (canceled)

8. (canceled)

10. (canceled)

11. (canceled)

12. (canceled)

14. (canceled)

15. (canceled)

16. (canceled)

17. (canceled)

18. (canceled)

19. (canceled)

20. (canceled)

21. (canceled)

22. (canceled)

23. (canceled)

24. (canceled)

27. (canceled)

28. (canceled)

30. (canceled)

31. (canceled)

32. (canceled)

33. (canceled)

34. (canceled)

35. (canceled)

36. (canceled)

37. (canceled)

38. (canceled)

39. (canceled)

40. (canceled)

41. (canceled)

42. (canceled)

43. (canceled)

44. (canceled)

45. (canceled)

46. (canceled)

47. (canceled)

48. (canceled)

53. (canceled)

54. (canceled)

56. (canceled)

57. (canceled)

58. (canceled)

59. (canceled)

60. (canceled)

61. (canceled)

62. (canceled)

63. (canceled)

64. (canceled)

65. (canceled)

66. (canceled)

67. (canceled)

68. (canceled)

69. (canceled)

70. (canceled)

72. (canceled)

73. (canceled)

76. (canceled)

77. (canceled)

78. (canceled)

79. (canceled)

82. (canceled)

83. (canceled)

85. (canceled)

86. (canceled)

87. (canceled)

88. (canceled)

89. (canceled)

90. (canceled)

92. (canceled)

93. (canceled)

94. (canceled)

95. (canceled)

102. (canceled)

103. (canceled)

108. (canceled)

109. (canceled)

110. (canceled)

114. (canceled)

116. (canceled)

117. (canceled)

118. (canceled)

119. (canceled)

120. (canceled)

122. (canceled)

123. (canceled)

125. (canceled)

133. (canceled)

135. (canceled)

136. (canceled)

138. (canceled)

140. (canceled)

142. (canceled)

145. (canceled)

147. (canceled)

149. (canceled)

150. (canceled)

151. (canceled)

152. (canceled)

153. (canceled)

154. (canceled)

155. (canceled)

159. (canceled)

162-326. -326. (canceled)

327. A computationally-implemented system, comprising

circuitry for receiving speech data correlated to one or more words spoken by a particular party;

circuitry for receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party;

circuitry for obtaining target data regarding a target configured to process at least a portion of the received speech data;

circuitry for determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data; and

circuitry for transmitting adaptation result data that is based on at least one aspect of the received speech data.

328. (canceled)

329. A device defined by a computational language comprising:

one or more interchained physical machines ordered for receiving speech data correlated to one or more words spoken by a particular party;

one or more interchained physical machines ordered for receiving adaptation data that is at least partly based on at least one speech interaction of a particular party that is discrete from the received speech data, wherein at least a portion of the adaptation data has been stored on a particular device associated with the particular party;

one or more interchained physical machines ordered for obtaining target data regarding a target configured to process at least a portion of the received speech data;

one or more interchained physical machines ordered for determining whether to apply the adaptation data for processing at least a portion of the received speech data, at least partly based on the acquired target data; and

one or more interchained physical machines ordered for transmitting adaptation result data that is based on at least one aspect of the received speech data.