Method and system for voice-enabled text entry

• US 7,369,988 B1
• Filed: 02/24/2003
• Issued: 05/06/2008
• Est. Priority Date: 02/24/2003
• Status: Active Grant

• Current Assignee: Sprint Spectrum LP (Deutsche Telekom AG)
• Original Assignee: Sprint Spectrum LP (Deutsche Telekom AG)
• Inventors: Heppner, Clyde C., Thenthiruperai, Balaji S.
• Primary Examiner(s): Hudspeth; David
• Assistant Examiner(s): Rider; Justin W.
• Application Number:
  US10/373,822
• Time in Patent Office: 1,898 Days
• Field of Search: 704/10
• US Class Current: 704/10
• CPC Class Codes: G10L 15/193 Formal grammars, e.g. finite state automata, context free grammars or word networks
• CPC Class Codes: G10L 2015/228 of application context
• CPC Class Codes: H04M 2250/70 methods for entering alphabetical characters, e.g. multi-tap or dictionary disambiguation methods for entering alphabetical characters per se G06F3/0237

Abstract

A method and system for voice enabled text entry. The method includes receiving a first user-entry with a communications device and creating a transient grammar set as a result of receiving the first user-entry, where the transient grammar set is based on the first user-entry. The method further includes providing an indication that the transient grammar set is created, the indication being provided using the communications device. The method still further includes receiving a second user-entry with the communications device and responsively selecting a text string from the transient grammar set, where the text string corresponds with the second user-entry.

Claims

1. A method of text entry comprising:

receiving a first user-entry with a communications device;

creating a transient grammar set as a result of receiving the first user-entry, the transient grammar set being based on the first user-entry, wherein creating the transient grammar set comprises selecting from a grammar store a subset of grammars that correspond to the first user-entry;

providing an indication that the transient grammar set is created, the indication being provided using the communications device;

receiving a second user-entry with the communications device;

responsively selecting a text string from the transient grammar set, the text string corresponding with the second user-entry; and

discarding the transient grammar set after the text string corresponding with the second user-entry has been selected.

2. The method of claim 1, wherein receiving the first user-entry comprises receiving one or more user-typed characters entered on the communications device.

3. The method of claim 2, further comprising transmitting the one or more user-typed characters from the communications device to a service node over a communications link.

4. The method of claim 3, wherein the communications link is a wireless communications link.

5. The method of claim 3, wherein transmitting the one or more user-typed characters comprises transmitting the one or more user-typed characters as a result of a threshold number of characters being entered on the communications device.

6. The method of claim 5, further comprising determining the threshold number of characters based on a first character of the user-typed characters.

7. The method of claim 5, wherein the threshold number of characters is determined by the communications device.

8. The method of claim 1, wherein selecting the subset of grammars comprises implementing a weighted statistical model using an artificial neural network software application.

9. The method of claim 1, wherein the step of providing an indication that the transient grammar set is created comprises generating an audio prompt using the communications device.

10. The method of claim 9, wherein the audio prompt is a text-to-speech prompt.

11. The method of claim 1, wherein providing an indication that the transient grammar set is created comprises displaying a list of text strings using the communications device, the list of text strings corresponding with the transient grammar set.

12. The method of claim 1, wherein receiving the second user-entry comprises receiving a spoken user-entry with the communications device.

13. The method of claim 12, wherein responsively selecting a text string that corresponds with the second user-entry comprises:

transmitting the spoken user-entry to a service node;

comparing the spoken user-entry with phonemes of the transient grammar set on the service node; and

matching the spoken user-entry with a text string that corresponds with a set of phonemes of the transient grammar set.

14. The method of claim 13, further comprising digitally sampling the spoken user-entry prior to transmitting the spoken user-entry to the service node.

15. The method of claim 12, wherein responsively selecting a text string that corresponds with the second user-entry comprises:

comparing the spoken user-entry with phonemes of the transient grammar set using the communications device; and

matching the spoken user-entry with a text string that corresponds with a set of phonemes of the transient grammar set.

16. The method of claim 1, wherein receiving the second user-entry comprises receiving a menu selection with the communications device.

17. The method of claim 16, wherein receiving a menu selection comprises receiving a signal generated by depressing a key on the communications device.

18. The method of claim 16, wherein receiving a menu selection comprises receiving a signal generated by mechanical contact with a touch-screen display of the communications device.

19. A text-messaging system comprising:

a communications device;

a communications service node communicatively coupled with the communications device, wherein at least one of the communications device and the service node, alone or in combination, implement a speech recognition engine;

a processor coupled with at least one of the communication device and the service node; and

a predictive engine executable by the processor for generating a transient grammar set based on a subset of grammars selected from a grammar store, the subset of grammars corresponding to a first user-entry made using the communications device;

wherein the speech recognition engine selects a text string associated with a portion of the transient grammar set based on a second user-entry made using the communications device, the second user-entry being a selection from the transient grammar set, and

wherein the transient grammar set is discarded after the text string associated with the portion of the transient grammar set has been selected.

20. The text-messaging system of claim 19, wherein the communications device comprises one of:

a wireless telephone;

a personal data assistant;

a laptop computer;

an Internet appliance, a desktop personal computer, and a wired telephone.

21. The text-messaging system of claim 19, wherein the communications device and the service node implement a distributed speech recognition engine.

22. The text-messaging system of claim 19, further comprising an electronic database coupled with the processor, the database including the grammar store.

23. The text-messaging system of claim 22, wherein the grammar store comprises grammars based on English language rules;

and

text messages communicated using the text-messaging system.

24. The text-messaging system of claim 22, wherein grammars associated with text messages communicated by a particular user are also associated with the particular user in the database.

25. The text-messaging system of claim 19, wherein the predictive engine comprises an artificial neural network.

26. The text messaging system of claim 25, wherein the artificial neural network comprises a weighted statistical model for predicting a desired text string based on the first user-entry.

27. The text-messaging system of claim 19, wherein, based on a mode selection on the communications device, the text string comprises one of a word, a phrase and a sentence.

28. The text-messaging system of claim 19, wherein the first entry is a typed text entry and the second entry is at least one of a spoken entry and a menu selection entry.

29. The text-messaging system of claim 19, further comprising a session manager for managing a text-messaging session, wherein the session manager synchronizes changing a communications link coupling the communications device and the service node between a voice channel and a data channel.

30. The text-messaging system of claim 29, wherein the session manager comprises logic to:

track communication protocol parameters for the data session and the voice session; and

responsively switch between the data session and the voice session based on an information type being received from the communications device.

31. A method for generating a text message comprising:

initiating a text-messaging application on a communications device;

establishing a real-time data session between the communications device and a service node;

receiving a first user-entry at the communications device, the first user-entry comprising one or more text characters corresponding with a desired text string;

transmitting the text characters from the communications device to the service node using the real-time data session;

generating a transient grammar set based on a subset of grammars selected from a grammar store at the service node, the subset of grammars corresponding to the first user-entry;

indicating that the transient grammar set is created using the communications device as a result of receiving a signal from the service node;

receiving a second user-entry at the communications device, the second user-entry corresponding with the desired text string and being a selection from a set of text strings corresponding with the transient grammar set;

transmitting the second user-entry from the communications device to the service node;

comparing the second user-entry with the transient grammar set to identify the desired text string, the comparison being done using the service node;

transmitting the desired text string from the service node to the communications device;

displaying the desired text string on the communications device in the text-messaging application; and

discarding the transient grammar set after the desired text string is identified.

32. The method of claim 31, wherein the text-messaging application comprises at least one of an instant messaging application, a short-message-system application, a short-mail application, and an email application.

33. The method of claim 31, wherein establishing a real time data session comprises establishing a real time protocol (RTP) session.

34. The method of claim 33, wherein the RTP session is established using session initiation protocol (SIP).

35. The method of claim 31, wherein receiving one or more text characters comprises receiving a predetermined number of user-entered text characters on the communications device, the user-entered text characters being entered using one or more keys included in the communications device.

36. The method of claim 31, wherein generating the transient grammar set comprises predicting, based on the predetermined number of text characters, the desired text string.

37. The method of claim 36, wherein predicting the desired text string is accomplished using an artificial neural network.

38. The method of claim 31, wherein indicating that the transient grammar set is created comprises playing a text-to-speech prompt on the communications device.

39. The method of claim 31, wherein indicating that the transient grammar set is created comprises displaying the set of text strings corresponding with the transient grammar set on the communications device.

40. The method of claim 31, wherein receiving the second user-entry at the service node comprises receiving digitized speech corresponding with a spoken user-entry from the communications device.

41. The method of claim 40, wherein the digitized speech comprises voice-over-packet data.

42. The method of claim 40, further comprising using the speech recognition engine to compare the digitized speech with phonemes included in the transient grammar set to determine the desired text string.