One-row keyboard and approximate typing

• US 20080270897A1
• Filed: 04/18/2008
• Published: 10/30/2008
• Est. Priority Date: 08/13/2004
• Status: Active Grant

• Current Assignee: 5 examples, Inc.
• Original Assignee: 5 examples, Inc.
• Inventors: Jawerth, Bjorn, Mehta, Viraj
• Granted Patent: US 8,147,154 B2
• Time in Patent Office: Days
• US Class Current: 715/261
• CPC Class Codes: G06F 1/162 changing, e.g. reversing, the face orientation of the screen with a two degrees of freedom mechanism, e.g. for folding into tablet PC like position or orienting towards the direction opposite to the user to show to a second user
• CPC Class Codes: G06F 1/1622 with enclosures rotating around an axis perpendicular to the plane they define or with ball-joint coupling, e.g. PDA with display enclosure orientation changeable between portrait and landscape by rotation with respect to a coplanar body enclosure
• CPC Class Codes: G06F 1/1666 Arrangements for reducing the size of the integrated keyboard for transport, e.g. foldable keyboards, keyboards with collapsible keys G06F1/1664 takes precedence
• CPC Class Codes: G06F 1/1671 Special purpose buttons or auxiliary keyboards, e.g. retractable mini keypads, keypads or buttons that remain accessible at closed laptop G06F1/1666 takes precedence
• CPC Class Codes: G06F 3/0221 Arrangements for reducing keyboard size for transport or storage, e.g. foldable keyboards, keyboards with collapsible keys G06F3/0216 takes precedence; for keyboards integrated in a laptop computer G06F1/1666
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: H01H 13/86 characterised by the casing, e.g. sealed casings or casings reducible in size
• CPC Class Codes: H01H 2223/05 composed of hingedly connected sections
• CPC Class Codes: H04M 1/021 using combined folding and rotation motions combined with translation movement H04M1/0235
• CPC Class Codes: H04M 1/0214 Foldable telephones, i.e. with body parts pivoting to an open position around an axis parallel to the plane they define in closed position combined with translation movement H04M1/0235
• CPC Class Codes: H04M 1/0225 Rotatable telephones, i.e. the body parts pivoting to an open position around an axis perpendicular to the plane they define in closed position combined with translation movement H04M1/0235
• CPC Class Codes: H04M 1/0235 Slidable or telescopic telephones, i.e. with a relative translation movement of the body parts; Telephones using a combination of translation and other relative motions of the body parts
• CPC Class Codes: H04M 1/0247 comprising more than two body parts
• CPC Class Codes: H04M 1/23 Construction or mounting of dials or of equivalent devices; Means for facilitating the use thereof by improving visibility H04M1/22
• CPC Class Codes: H04M 2250/18 including more than one keyboard unit

Abstract

In one aspect, the present invention comprises an apparatus for character entry on an electronic device, comprising: a keyboard with one row of keys; and an electronic display device in communication with the keyboard; wherein one or more keys on the keyboard has a correspondence with a plurality of characters, and wherein the correspondence enables QWERTY-based typing. In another aspect, the invention comprises an apparatus for character entry on an electronic device, comprising: a keyboard with a plurality of keys; and an electronic display device in communication with the keyboard; wherein one or more keys on the keyboard has a correspondence with a plurality of characters, and wherein, for each of the one or more keys, the plurality of characters comprises: (a) a home row character associated with a particular finger when touch typing; and (b) a non-home-row character associated with the particular finger when touch typing.

Claims

1. A method for predicting a character sequence based on a signature sequence of equal length to that of said character sequence, said signature sequence generated by a user'"'"'s consecutive selection of keys on a reduced keyboard, comprising:

(i) receiving a signature for each key selected by said user, each key-signature pair of said reduced keyboard being unique, each signature associated with at least one dual signature and at least one of said signatures associated with two or more dual signatures, each signature-dual signature pair uniquely associated with a character supported by said reduced keyboard;

(ii) storing in a computer readable memory a plurality of objects, each of said objects comprising;

(a) a unique sequence of signature-dual signature pairs derived from at least one set of legal beginning-of-word (BOW) character combinations; and

(b) one or more attributes associated with said sequence;

(iii) for each received signature, predicting at least one of said plurality of objects based on a sequence of said signatures, including said received signature, consecutively received from said reduced keyboard and on said object sequence attributes;

each such sequence terminating with a signature predicted as a delimiter; and

(iv) outputting a unique BOW character combination associated with said at least one predicted object.

2. The method of claim 1, wherein at least one of said signature-dual signature pairs is associated with a letter character and at least one different signature-dual signature pair with the same signature is associated with a non-letter character.

3. The method of claim 1, further comprising combining two or more of said sets of legal BOW character combinations to create an extended set of legal BOW character combinations.

4. The method of claim 3, wherein said extended set of legal BOW character combinations includes punctuation characters.

5. The method of claim 1, further comprising predicting at least one of said objects based on specific rules of disambiguation.

6. The method of claim 5, wherein said rules of disambiguation are rules for punctuation.

7. The method of claim 1 wherein for each of said plurality of objects at least one of said object sequence attributes is path frequency.

8. The method of claim 1, further comprising outputting said predicted objects in an order based on said object sequence attributes.

9. The method of claim 8, wherein said object sequence attributes includes at least path frequency and said predicted objects output order is based, as least in part, on said path frequency.

10. The method of claim 1, wherein said prediction is based, at least in part, on dual signature selections resulting from a Display Optimization process.

11. The method of claim 1, further comprising locking said dual signature associated with at least one of said signature-dual signature pairs in said sequence of signatures, and further basing said prediction, at least in part, on said at least one locked dual signature.

12. The method of claim 1, wherein for each of said plurality of objects at least one of said object sequence attributes is associated with at least one context tag, and said at least one context tag is used, at least in part, to predict at least one of said objects.

13. The method of claim 12, wherein at least one of said context tags is associated with a part-of-speech tag.

14. The method of claim 1, wherein said memory contains only objects whose sequence of signature-dual signature pairs is either (a) a delimiter-terminated sequence of signatures or (b) a sequence of signatures less than or equal to some fixed length.

15. The method of claim 14, wherein, on an as needed basis, objects comprising sequences of signatures greater than said fixed length are derived from at least one object containing a delimiter-terminated sequence of signatures comprising the same initial sequence of signatures.

16. The method of claim 3, further comprising rules to generate valid delimiter-terminated character combinations from other valid delimiter-terminated character combinations in said set of legal BOW character combinations.

17. The method of claim 16, wherein said rules include rules for generating standard modifications, endings, and punctuation of basic words within a specific context in said set of legal BOW character combinations.

18. The method of claim 3, wherein at least one of said sets of legal character combinations is derived from an address book or list of email addresses.

19. The method of claim 1, wherein said legal BOW character combinations are defined from terms or symbols used within one or more specific contexts.

20. The method of claim 1, wherein said received signature comprises one or more, but less than ‘

n’

, signatures of an n-tuple signature, each n-tuple signature uniquely associated with a character supported by said reduced keyboard; and

wherein each of said objects comprises a unique sequence of n-tuple signatures derived from said at least one set of legal BOW character combinations.

21. The method of claim 1, wherein one or more of said received signatures are generated by two or more keys being simultaneously selected by said user.

22. A method for predicting a character sequence based on a signature sequence generated by a user'"'"'s consecutive selection of keys on a reduced keyboard, wherein said signature sequence is an approximation to a signature sequence associated with said character sequence, comprising:

(i) receiving a signature for each key selected by said user, each key-signature pair of said reduced keyboard being unique, each signature associated with at least one dual signature and at least one signature associated with one or more dual signatures, each signature-dual signature pair associated with a unique character supported by said reduced keyboard;

(ii) storing in a computer readable memory a plurality of objects, each of said objects comprising;

(a) a unique sequence of signature-dual signature pairs derived from at least one set of legal character combinations; and

(b) one or more attributes associated with said sequence;

(iii) predicting at least one approximate signature sequence based a sequence of signatures consecutively received from said reduced keyboard;

(iv) for each of said approximate signature sequences predicting at least one of said plurality of objects based on said approximate signature sequence and on said object sequence attributes, terminating each such approximate signature sequence when said approximate signature sequence terminates with a signature predicted as a delimiter; and

(v) outputting a character combination associated with at least one of said predicted plurality of objects.

23. The method of claim 22, wherein a prediction of an approximate signature sequence is based on, at least in part, one or more similarity measures or typing models.

24. The method of claim 23, wherein at least one of said one or more similarity measures or typing models is based on, at least in part, observed user key selection behaviors.

25. The method of claim 24, wherein said observed user key selection behaviors are adjusted to account for behaviors of a specific user.

26. The method of claim 24, wherein said observed user key selection behaviors are adjusted to account for a specific context or environment.

27. The method of claim 22, wherein a prediction of at least one of said plurality of objects is based on, at least in part, a reduced keyboard disambiguation method and an approximate signature sequence.

28. The method of claim 27, wherein said reduced keyboard disambiguation method comprises:

(vi) receiving at least one approximate signature sequence;

(vii) storing in a computer readable memory a plurality of objects, each of said objects comprising;

(a) a unique sequence of signature-dual signature pairs derived from at least one set of legal beginning-of-word (BOW) character combinations; and

(b) one or more attributes associated with said sequence;

(viii) for each received approximation signature sequence, predicting at least one of said plurality of objects based on said approximation signature sequence and on said object attributes; and

(ix) outputting a unique BOW character combination associated with said at least one predicted object.

29. The method of claim 27, wherein said object predicting step is performed for only delimiter-terminated approximate signature sequences.

30. The method of claim 27, wherein each sequence of signature-dual signature pairs corresponding to each of said plurality of objects is a delimiter-terminated sequence.

31. The method of claim 22, further comprising outputting said predicted objects in an order based on said object sequence attributes.

32. The method of claim 31, wherein at least one of said object sequence attributes is frequency-of-use.

33. The method of claim 32, wherein said frequency-of-use is based on a path-frequency.

34. The method of claim 22, further comprising applying a cost function which assigns a cost to each of said approximate signature sequences.

35. The method of claim 34, wherein said cost function is based on one or more edit distances, pseudo-distances, or sequence transformations.

36. The method of claim 34, further comprising outputting said predicted objects in an order based on, at least in part, one or more of said cost function and said object sequence attributes.

37. The method of claim 34, wherein:

(a) said cost is less than some fixed value;

(b) at least one of said object attributes is a path-frequency; and

(c) said path-frequency is greater than some minimum value greater than zero.

38. The method of claim 22, wherein one or more of said signatures in an approximate signature sequence is locked to a value and the corresponding one or more signatures in each approximate signature sequence is locked to the same value.

39. The method of claim 38, wherein said one or more locked signatures and corresponding values are determined, at least in part, by a display optimization process, and further comprising basing said prediction, at least in part, on said locked signatures.

40. The method of claim 38, further comprising locking a dual signature associated with at least one of said one or more locked signatures, and further basing said prediction, at least in part, on said locked dual signature.

41. The method of claim 22, wherein at least one of said object sequence attributes is associated with at least one context tag, and said at least one context tag is used, at least in part, to predict at least one of said objects.

42. The method of claim 41, wherein at least one of said context tags is associated with a part-of-speech.

43. The method of claim 22, wherein said memory contains only objects whose sequence of signature-dual signature pairs is either (a) a delimiter-terminated sequence of signatures, or (b) a sequence of signatures less than or equal to some fixed length.

44. The method of claim 43, wherein, on an as needed basis, objects comprising sequences of signatures greater than said fixed length are derived from at least one object containing a delimiter-terminated sequence of signatures comprising the same initial sequence of signatures.

45. The method of claim 22, further comprising rules to generate valid delimiter-terminated character combinations from other valid delimiter-terminated character combinations in said set of legal character combinations.

46. The method of claim 45, wherein said rules include rules for generating standard modifications, endings, and punctuation of basic words within a specific context in said set of legal character combinations.

47. The method of claim 22, wherein said legal character combinations are defined from terms or symbols used within one or more specific contexts.

48. The method of claim 47, wherein said specific contexts comprise an address book or email addresses.

49. The method of claim 22, wherein said received signature comprises one or more, but less than ‘

n’

, signatures of an n-tuple signature, each n-tuple signature uniquely associated with a character supported by said reduced keyboard; and

wherein each of said objects comprise a unique sequence of n-tuple signatures derived from said at least one set of legal character combinations.

50. The method of claim 22, wherein one or more of said received signatures are generated by two or more keys being simultaneously selected by a user of said reduced keyboard.

51. A method for reducing predictive uncertainty in a disambiguation group, comprising:

(i) receiving and storing in a computer readable medium a disambiguation group comprising two or more objects, each of said objects associated with one or more object equivalence classes and a legal combination of characters supported by a reduced keyboard;

(ii) receiving and storing in a computer readable medium one or more rules defining legal object equivalence classes or relationships between object equivalence classes;

(iii) predicting at least one object from said disambiguation group based on said object equivalence class association of each of said disambiguation group objects and said one or more rules defining legal object equivalence classes or relationships between object equivalence classes; and

(iv) outputting said at least at least one predicted object from said disambiguation group.

52. The method of claim 51, wherein each of said objects is further associated with a legal character combination and a length of said legal character combination is (a) greater than some minimum value greater than one, or (b) is delimiter-terminated.

53. The method of claim 52, wherein said legal character combinations comprise only delimiter-terminated character combinations.

54. The method of claim 51, wherein said disambiguation group is determined by the steps of:

(v) receiving a signature for each key selected by a user of a reduced keyboard, each key-signature pair of said reduced keyboard being unique, each signature associated with at least one dual signature and at least one signature associated with one or more dual signatures, each signature-dual signature pair associated with a unique character supported by said reduced keyboard;

(vi) storing in a computer readable memory a plurality of objects, each of said objects comprising;

(a) a unique sequence of signature-dual signature pairs derived from at least one set of legal character combinations; and

(b) one or more attributes associated with said sequence;

(vii) predicting at least one approximate signature sequence based a sequence of signatures consecutively received from said reduced keyboard

(viii) for each said approximate signature sequence predicting at least one of said plurality of objects based on said approximate signature sequence and on said object attributes, terminating each such approximate signature sequence when said approximate signature sequence terminates with a signature predicted as a delimiter; and

(ix) outputting at least one of said predicted plurality of objects as a disambiguation group.

55. The method of claim 51, wherein said disambiguation group is determined by a reduced keyboard disambiguation method.

56. The method of claim 55, wherein said reduced keyboard disambiguation method comprises:

(v) receiving a signature for each key selected by a user of a reduced keyboard, each key-signature pair of said reduced keyboard being unique, each signature associated with at least one dual signature and at least one of said signatures associated with two or more dual signatures, each signature-dual signature pair uniquely associated with a character supported by said reduced keyboard;

(vi) storing in a computer readable memory a plurality of objects, each of said objects comprising;

(a) a unique sequence of signature-dual signature pairs derived from at least one set of legal beginning-of-word (BOW) character combinations; and

(b) one or more attributes associated with said sequence;

(vii) for each received signature, predicting at least one of said plurality of objects based on a sequence of said signatures, including said received signature, consecutively received from said reduced keyboard and on said object attributes;

each such sequence terminating with a signature predicted as a delimiter; and

(viii) outputting at least one of said predicted plurality of objects as a disambiguation group.

57. The method of claim 51, wherein said rules include rules based on commonly shared object equivalence classes.

58. The method of claim 57, wherein at least one of said commonly shared object equivalence classes defines membership in a common dictionary.

59. The method of claim 58, wherein said common dictionary is a BOW dictionary.

60. The method of claim 57, wherein said commonly shared object equivalence class defines specific functions or operations.

61. The method of claim 57, wherein at least one of said commonly shared object equivalence classes defines specific styles.

62. The method of claim 51, further comprising:

(v) receiving and storing in a computer readable medium one or more delimiter-terminated objects, each of said delimiter-terminated objects being associated with user inputs received prior to or subsequent to inputs associated with said disambiguation group, and each of said delimiter-terminated objects associated with one or more object equivalence classes; and

(vi) predicting at least one object from said disambiguation group based, at least in part, on said equivalence classes of said received delimiter-terminated objects.

63. The method of claim 62, further comprising applying rules defining legal combinations of object-object equivalence pairs.

64. The method of claim 63, wherein said rules are based, at least in part, on rules of Natural Language Processing.

65. The method of claim 64, wherein said Natural Language Processing is based on grammar rules.

66. The method of claim 65, wherein said grammar rules comprise rules relating to parts-of-speech.

67. The method of claim 63, wherein said rules are based on probabilistic inference of terminating object-object equivalence class pairs in legal combinations of object-object equivalence class pairs.

68. The method of claim 67, wherein said probabilistic inference is learned from a fixed source.

69. The method of claim 67, wherein said probabilistic inference is learned on-the-fly based on user input.

70. The method of claim 67, wherein said rules admit said object-object equivalence pairs within said disambiguation group only with a probabilistic inference exceeding some minimum value greater than zero.

71. The method of claim 67, wherein said rules admit said object-object equivalence pairs within said disambiguation group only with a probabilistic inference greater than the frequency-of-use of the object with the greatest frequency-of-use in the disambiguation group.

72. The method of claim 67, wherein each of said objects in said disambiguation group is further associated with a frequency-of-use.

73. The method of claim 72, wherein said frequency-of-use is path frequency.

74. The method of claim 72, wherein said prediction is based, at least in part, on selecting an object in said disambiguation group with a greater of either frequency-of-use or probabilistic inference.

75. The method of claim 62, wherein in said step of receiving and storing in a computer readable medium one or more delimiter-terminated objects, each of said delimiter-terminated objects is only associated with user inputs received prior to, and not subsequent to, inputs associated with said disambiguation group.

76. The method of claim 75, wherein at least one of said delimiter-terminated objects is a delimiter-terminated object associated with user input received just prior to inputs associated with said disambiguation group.

77. The method of claim 75, wherein said delimiter-terminated objects comprise two delimiter-terminated objects associated with user input received just prior to inputs associated with said disambiguation group.

78. The method of claim 51, wherein said object equivalence classes are generated by aggregating more than one object equivalence class from a first set of object equivalence classes into a new object equivalence class.

79. A method for selecting a character combination in a disambiguation group for output to a user comprising:

(i) receiving a disambiguation group of two or more objects, each of said objects associated with a legal combination of characters supported by a reduced keyboard;

(ii) implementing a display algorithm that selects one object in said disambiguation group based on one or more quantifications of user experience;

(iii) outputting a character combination associated with said selected object for display to said user.

80. The method of claim 79, wherein at least one of said one or more quantifications of user experience is based, at least in part, on minimizing one or more cost functions.

81. The method of claim 80, wherein said objects are further associated with a frequency-of-use.

82. The method of claim 81, wherein minimizing cost is based on (a) minimizing average cost associated with said one or more cost functions, (b) said character combinations, and (c) said frequency-of-use associated with each of said objects.

83. The method of claim 80, wherein at least one cost function is based on a differential or incremental cost function.

84. The method of claim 80, wherein cost associated with at least one cost function is based, at least in part, on a number of characters associated with said objects in said disambiguation group.

85. The method of claim 79, wherein said display algorithm is based on, at least in part, a beginning of word path ordering (BOW/po).

86. The method of claim 79, wherein said display algorithm is based on, at least in part, minimizing number of changes to a character combination currently displayed to said user.

87. The method of claim 79, further comprising:

(iv) receiving a signal from said reduced keyboard indicating that the user has rejected said character combination associated with said selected object;

(v) modifying said disambiguation group by eliminating said object associated with said rejected character combination;

(vi) modifying said one or more quantifications of user experience based on said eliminated object and selecting one object from said modified disambiguation group based on said modified one or more quantifications of user experience;

(vii) outputting said character combination associated with said selected object for display to said user.