Portable wire-less communication device

• US 20050131687A1
• Filed: 09/24/2004
• Published: 06/16/2005
• Est. Priority Date: 09/25/2003
• Status: Abandoned Application

• Current Assignee: Canon Europa N.V. (Canon Inc.)
• Original Assignee: Canon Europa N.V. (Canon Inc.)
• Inventors: Sorrentino, Andrea
• Application Number:
  US10/948,263
• Publication Number:
  US 20050131687A1
• Time in Patent Office: Days
• US Class Current: 704/235
• CPC Class Codes: G10L 15/28 Constructional details of speech recognition systems
• CPC Class Codes: H04M 1/271 controlled by voice recognition
• CPC Class Codes: H04M 1/72436 for text messaging, e.g. short messaging services [SMS] or e-mails

Abstract

A cellular telephone is described which includes a predictive text editor for generating text messages in response to key-presses made on an ambiguous keyboard of the cellular telephone. The text editor also includes a speech recogniser for recognising words in speech input by the user to disambiguate between possible words corresponding to key-presses made by the user on the ambiguous keyboard.

Claims

1. A portable wire-less communication device comprising:

a plurality of keys for the input of symbols, wherein each of at least some of the keys is operable for the input of a plurality of different symbols;

a keyboard processor operable to generate text data in dependence upon the actuation of one or more of said keys by a user;

an automatic speech recogniser operable to recognise an input speech signal and to generate a recognition result; and

a controller responsive to the text data generated by said keyboard processor and responsive to said recognition result generated by said automatic speech recogniser to generate text.

2. A device according to claim 1, wherein said automatic speech recogniser includes a vocabulary which defines the possible words that can be recognised by the speech recogniser and wherein said speech recogniser is responsive to text data generated by the keyboard processor to restrict the speech recognition vocabulary prior to recognition processing of said speech signal.

3. A device according to claim 1, wherein said plurality of keys operable for the input of the plurality of different symbols form part of an ambiguous keyboard.

4. A device according to claim 2, wherein said keyboard processor is a predictive text editor.

5. A device according to claim 4, wherein said keyboard processor is operable, in response to actuation of said keys, to generate text data that defines predicted symbols intended by the user and operable to regenerate text data that defines re-predicted symbols in response to further key actuation.

6. A device according to claim 5, wherein said speech recogniser is operable to recognise said speech signal in dependence upon at least one of the predicted symbols defined by said text data generated by said keyboard processor and is operable, in response to a regeneration of a said text data by said keyboard processor, to re-perform speech recognition on the speech signal in dependence upon at least one of the predicted symbols defined by the re-generated text data.

7. A device according to claim 5, wherein said keyboard processor is operable to receive a key ID identifying a latest key pressed by the user and is operable to store previous key-press data indicative of the input key sequence for a current word being entered via the keys.

8. A device according to claim 7, further comprising a text graph which defines a mapping between previous key-press data and a latest key ID to text data identifying the most likely word corresponding to the input key sequence, and wherein said keyboard processor is operable to use the key ID for the latest key press and the stored previous key-press data to address said text graph to determine the text data identifying the most likely word corresponding to the input key sequence.

9. A device according to claim 8, wherein said text graph also defines a mapping between said previous key data and said latest key ID to data identifying possible words corresponding to the input key sequence and wherein said automatic speech recogniser is responsive to the data identifying possible words corresponding to an input key sequence to restrict the recognition process thereof.

10. A device according to claim 9, wherein said keyboard processor is operable to address said text graph using said previous key-press data and the current key ID to retrieve the data identifying possible words corresponding to the input key sequence and is operable to pass the data identifying the possible words to said automatic speech recogniser

11. A device according to claim 10, wherein said automatic speech recogniser is operable to restrict a vocabulary thereof in dependence upon the data identifying said possible words received from said keyboard processor.

12. A device according to claim 9, comprising a word dictionary having N word entries, each storing word data for a word, wherein the word entries are ordered in the word dictionary based on the input key sequence needed to enter the symbols for the word via said keys, wherein each word entry has an associated index value indicative of the order of the word entry in the dictionary, and wherein the text data identifying the most likely word comprises the index value of that word in said word dictionary.

13. A device according to claim 12, wherein said text data identifying possible words corresponding to the input key sequence comprises the index value for at least one word in the dictionary and a range of index values for words in the dictionary that are adjacent to said at least one word in the dictionary.

14. A device according to claim 13, wherein said text data identifying possible words comprises the index value for the first or last of the possible words within the dictionary and the number of words appearing immediately after or before the identified first or last word.

15. A device according to claim 1, wherein said controller is operable to activate said automatic speech recogniser in response to speech received by the user and is operable to reactivate the speech recogniser in response to updated text data received from said keyboard processor.

16. A device according to claim 1, wherein said automatic speech recogniser comprises a grammar which defines all possible words that can be recognised by the speech recogniser and model data for the words.

17. A device according to claim 16, wherein said model data comprises subword unit models and wherein said grammar defines a sequence of subword unit models for each word.

18. A device according to claim 17, wherein said model data comprises phoneme-based models.

19. A device according to claim 18, wherein said model data comprises a mixture of tri-phone and bi-phone models for one or more words in the grammar.

20. A device according to claim 16, further comprising an activation unit operable to enable or disable portions of the grammar selected in accordance with text data generated by said keyboard processor in response to actuation of said keys by the user

21. A device according to claim 1, further comprising a word dictionary comprising N word entries, each storing word data for a word, wherein the word entries are ordered in the word dictionary based on the input key sequence needed to enter the symbols for the word using said keys and wherein said automatic speech recogniser is operable to recognise said word in dependence upon the data stored in said word dictionary.

27. A method according to claim 26, wherein the method is performed on a portable wire-less communication device according to any one of claims 1, 22, 24 and 25.

40. A storage medium storing computer program instructions for programming a portable wire-less communication device to become configured as a device according to any one of claims 1 to 25.

41. A physically-embodied computer program product carrying computer program instructions for programming a portable wire-less communication device to become configured as a device according to any one of claims 1 to 25.

22. A portable wire-less communication device, comprising:

a keypad having a plurality of keys for the input of symbols, wherein each of at least some of the keys is operable for the input of a plurality of different symbols;

a text message generator responsive to keypad input to generate text for a text message; and

a speech recogniser responsive to voice input to determine a spoken word;

wherein;

the text message generator is responsive to the determination of a word by the speech recogniser to include the word in the text message.

23. A device according to claim 22, wherein the speech recogniser is operable to determine a word in dependence upon at least part of the content of the text message entered via the keypad.

24. A portable wire-less communication device, comprising:

a keypad having a plurality of keys for the input of symbols, wherein each of at least some of the keys is operable for the input of a plurality of different symbols;

a text message generator responsive to keypad input to generate text for a text message; and

a speech recogniser responsive to voice input to determine a spoken word;

wherein;

the speech recogniser is operable to determine a word in dependence upon at least part of the content of the text message entered via the keypad.

25. Apparatus for generating and sending text messages over a communication network, the apparatus comprising:

a plurality of keys for the input of symbols, wherein the number of keys is less than the number of symbols;

a predictive text generator responsive to actuation of the keys to predict symbols intended by the user and to add the symbols to a text message, and operable to re-predict symbols in response to further key actuation and to change the symbols in the text message in accordance with the re-prediction; and

a speech recogniser operable to generate text for the text message by;

recognising a word spoken by a user, such that the recognition is performed in dependence upon at least one symbol generated by the predictive text generator;

storing in memory the voice data of the word spoken by the user; and

in response to re-prediction of a symbol by the predictive text generator, re-performing speech recognition using the stored voice data and in dependence upon the re-predicted symbol.

26. A method of generating text on a portable wire-less communication device having a plurality of keys for the input of symbols, wherein each of at least some of the keys is operable for the input of a plurality of different symbols, the method comprising:

generating text data in dependence upon the actuation of one or more of said keys by a user;

using an automatic speech recogniser to recognise an input speech signal to generate a recognition result; and

generating text in dependence upon text data generated by the actuation of said one or more keys by the user and in dependence upon the recognition result generated by said speech recogniser.

28. A data processing method comprising:

receiving text data representative of text for a plurality of words;

receiving mapping data defining a mapping between key-presses of an ambiguous keyboard and text symbols;

processing the text data and the mapping data to determine a key sequence for each word which defines the sequence of key-presses on said ambiguous keyboard which map to the text symbols corresponding to the word; and

sorting the respective text data for said plurality of words based on the key sequence determined for each word, to generate word dictionary data for use in an electronic device having such an ambiguous keyboard.

29. A method according to claim 28, wherein said sorting process orders the respective text data for each word based on an assigned order given to the keys of the ambiguous keyboard.

30. A method according to claim 29, wherein the keys of said ambiguous keyboard are assigned a numerical order and wherein said sorting process sorts the text data for each word based on the numerical order of each key sequence.

31. A method according to claim 28, further comprising a process of generating a signal carrying said word dictionary data.

32. A method according to claim 31, further comprising a process of recording said signal directly or indirectly on a recording medium.

33. A method according to claim 28, further comprising a process of processing said word dictionary data to generate data defining a predictive text graph which relates an input key sequence to data defining all words within said dictionary whose key sequence starts with said input key sequence.

34. A method according to claim 33, wherein said process of processing said word dictionary data generates data defining a predictive text graph which relates an input key sequence to data defining a most likely word corresponding to said input key sequence.

35. A method according to claim 33, further comprising a process of generating a signal carrying said data defining the predictive text graph.

36. A method according to claim 35, further comprising a process of recording said signal directly or indirectly on a recording medium.

37. A data processing method comprising:

receiving text data representative of text for a plurality of words;

receiving mapping data defining a mapping between key-presses of an ambiguous keyboard and text symbols;

processing the text data and the mapping data to determine a key sequence for each word which defines the sequence of key-presses on said ambiguous keyboard which map to the text symbols which correspond to the word;

receiving ASR grammar data identifying portions of the ASR grammar corresponding to each of said plurality of words; and

associating the determined key sequence for a word with the corresponding ASR grammar data for that word, to generate word dictionary data for use in an electronic device having such an ambiguous keyboard.

38. A method according to claim 37, further comprising a process of generating a signal carrying said word dictionary data.

39. A method according to claim 38, further comprising a process of recording said signal directly or indirectly on a recording medium.