Method and apparatus for text input in various languages

• US 7,636,083 B2
• Filed: 02/16/2005
• Issued: 12/22/2009
• Est. Priority Date: 02/20/2004
• Status: Expired due to Fees

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Tegic Communications, Inc. (Microsoft Corporation)
• Inventors: Aoki, Ann Naomi, James, Christina, Van Meurs, Pim
• Primary Examiner(s): Nguyen; Chanh
• Assistant Examiner(s): Mistry; Ram A
• Application Number:
  US11/060,012
• Publication Number:
  US 20050195171A1
• Time in Patent Office: 1,770 Days
• Field of Search: 434/319, 434/185, 341/28, 341/20, 341/22, 400/110, 400/100, 400/489, 400/472, 400/484, 400/486, 455/564, 715/535, 715/526, 715/530, 704/8, 704/260, 704/235, 345/171, 345/168, 345/172, 707/101, 707/102, 379/368
• US Class Current: 345/172
• CPC Class Codes: G06F 2203/0381 Multimodal input, i.e. interface arrangements enabling the user to issue commands by simultaneous use of input devices of different nature, e.g. voice plus gesture on digitizer
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 3/038 Control and interface arrangements therefor, e.g. drivers or device-embedded control circuitry

Abstract

A method and apparatus for facilitating text input for network applications, such as, for example, wireless applications, are described. Multiple input signals are received from a user, each input signal corresponding to a key within a user input interface and further representing Latin, e.g. English, language phonetic equivalents of at least one language character including a language character intended by the user to form a word construct in a language. Each input signal is further used to generate at least one proposed language character. At least one proposed word construct is further formed in the language using the at least one proposed language character. Finally, the at least one proposed word construct is displayed for the user on a display screen.

Claims

1. A method, comprising the steps of:

mapping each character or phonetic symbol in a language to a key on a reduced keyboard by associating its corresponding sound with the closest phonetic sound corresponding to a Latin-based letter assigned to a reduced keyboard key;

entering a series of constituent base characters of a desired word in a phonetic order using said mapped phonetic symbols, wherein said step of entering comprises;

receiving at least a first input signal representing a Latin phonetic equivalent of at least one character;

receiving a switching signal indicating a mode change followed by receiving at least one additional input signal in a particular position within the series of constituent base characters, wherein said at least one additional input signal is translated into a Hindu, Bengali, or Tamil character, forming a specific language character intended for a particular position; and

forming a list of possible word constructs containing both (1) said first input signal representing a Latin phonetic equivalent of at least one character and (2) said at least one additional input signal represented as a Hindu, Bengali, or Tamil character, wherein said list possible word constructs includes possible word constructs containing said constituent base characters in an order different than said phonetic order;

filtering possible word constructs not containing said specific language character in the intended position from said list of possible word constructs, forming a filtered list of word constructs; and

displaying said filtered list of word constructs.

2. The method of claim 1, wherein if said character or phonetic symbol represents a dipthong of more than one sound, a first sound of said dipthong is used to determine said mapping.

3. A text input method, comprising the steps of:

receiving a plurality of input signals from a user, each input signal corresponding to a key on a reduced keyboard within a user input interface and further representing phonetic equivalents of at least one language character including a language character intended by said user to form a word construct in a language, wherein said step of receiving comprises;

receiving a first input signal representing a Latin phonetic equivalent of at least one character;

receiving at least one additional input signal at a particular position within the plurality of input signals, following receiving a switching signal indicating a mode change, wherein said at least one additional input signal is translated into a Hindu, Bengali, or Tamil character, forming a specific language character intended for a particular position; and

forming a list of possible word constructs in said language using both (1) said first signal representing a Latin phonetic equivalent of at least one character and (2) said at least one additional input signal represented as a Hindu, Bengali, or Tamil character;

filtering possible word constructs not containing said specific language character in the intended position from said list of possible word constructs, forming a filtered list of word constructs; and

displaying said filtered list of word constructs for said user on a display screen.

4. The method according to claim 3, wherein said forming step further comprises the step of:

matching said input signals with a corresponding entry in a language database.

5. The method according to claim 3, further comprising the step of:

facilitating selection by said user of at least one of said possible word constructs displayed on said display screen.

6. The method of claim 3, further comprising the steps of:

forming a plurality of character sequences based on partial and/or completed word constructs retrieved from a database.

7. The method according to claim 3, further comprising the steps of:

facilitating selection by said user of a word construct that partially or most closely matches said word construct from said at least one proposed word construct; and

facilitating input of further input signals from said user to form said word construct.

8. A text entry method, comprising the steps of:

receiving a plurality of input signals from a user, each of said input signals corresponding to a key on a reduced keyboard within a user input interface and further representing language phonetic equivalents of at least one language character including a language character intended by said user to form a word construct in a language, wherein said step of receiving comprises;

receiving a first input signal representing a Latin phonetic equivalent of at least one character;

receiving a switching signal indicating a mode change followed by receiving at least one additional input signal in a particular position, wherein said at least one additional second input signal is translated into a Hindu, Bengali, or Tamil character, forming a specific language character intended for a particular position; and

using said first input signal to generate a first proposed language character;

using said at least one additional input signal to generate a at least one additional proposed language character;

displaying successively said at first proposed language character and said at least one additional proposed language character for said user on a display screen;

facilitating successive selection by said user of said language character for said each input signal to form said word construct; and

forming a list of possible word constructs using both (1) said first input signal representing a Latin phonetic equivalent of at least one character and (2) said second input signal represented as a Hindu, Bengali, or Tamil character;

filtering possible word constructs not containing said specific language character in the intended position from said list of possible word constructs, forming a filtered list of word constructs.

9. The method of claim 8, further comprising the step of:

matching and displaying word completions based on initial or partial word constructs.

10. A text entry apparatus, comprising:

a user input interface for receiving a plurality of input signals from a user, each input signal corresponding to a key on a reduced keyboard within said user input interface and further representing language phonetic equivalents of more than one language character including a language character intended by said user to form a word construct in a language, wherein said plurality of input signals comprise all of;

at least one ambiguous input signal representing a Latin phonetic equivalent of at least one character;

a switching signal indicating a mode change to an explicit mode of character entry whereby specific language characters are entered through the input of a key by a means for disambiguating the intended entry of the more than one language character present in the inputted key; and

at least one additional input signal comprising an explicit character at a particular position within the plurality of input signals, wherein said at least one additional input signal is translated into a Hindu, Bengali, or Tamil character, forming a specific language character intended for a particular position; and

forming a list of possible word constructs using both (1) said at least one ambiguous signal representing a Latin phonetic equivalent of at least one character and (2) said at least one additional input signal represented as a Hindu, Bengali, or Tamil character;

a processor coupled to said user input interface for using said each ambiguous input signal to generate at least one proposed language character, for forming a list of proposed word constructs in said language using said at least one proposed language character, and for filtering said proposed word constructs not containing said explicit character at said particular position within the plurality of input signals, forming a filtered list of word constructs; and

a display engine coupled to said processor for displaying said filtered list of word constructs for said user on a display screen.

11. The apparatus according to claim 10, wherein said processor further comprises:

means for retrieving said list of proposed word constructs from a language database.

12. The apparatus according to claim 10, wherein said user input interface further comprising:

means for facilitating selection by said user of said word construct from said filtered list of word constructs displayed on said display screen.

13. The apparatus according to claim 10, wherein said user input interface further comprising:

means for facilitating selection by said user of a word from the filtered list of word constructs.

14. A text entry apparatus, comprising:

a user input interface for receiving a plurality of input signals from a reduced keyboard by a user wherein said input signals comprise all of;

a first input signal representing a Latin phonetic equivalent of at least one character; and

a second input signal following a switching signal indicating a mode change, wherein said second input signal is translated into a Hindu, Bengali, or Tamil character, forming a specific language character intended for a particular position; and

a processor coupled to said user input interface for using said each input signal to generate a plurality of proposed language characters; and

a display engine coupled to said processor for displaying successively said plurality of proposed language characters for said user on a display screen, forming a list of possible word constructs;

a means for filtering possible word constructs not containing said proposed language characters in the intended positions from said list of possible word constructs, forming a filtered list of word constructs containing potential word constructs; and

said user input interface further comprising means for facilitating selection by the user of a potential word construct.

15. A text entry system, comprising:

means for receiving a plurality of input signals from a reduced keyboard by a user, each input signal corresponding to a key within a user input interface and further representing language phonetic equivalents of at least one language character comprising a language character intended by said user to form a word construct in a language, wherein said input signals comprise all of;

a first input signal representing a Latin phonetic equivalent of at least one character; and

at least one additional input signal in a particular position, following a switching signal indicating a mode change, wherein said at least one additional input signal is translated into a Hindu, Bengali, or Tamil character, forming a specific language character intended for a particular position; and

means for using said each input signal to generate a list of possible word constructs containing at least one proposed language character, wherein said at least one proposed language character comprises a consonant with a declined implicit vowel;

a means for filtering possible word constructs not containing said at least one additional input signal in the intended position from said list of possible word constructs, forming a filtered list of word constructs;

means for displaying successively said filtered list of word constructs for said user on a display screen; and

means for facilitating successive selection by said user of said language character for said each input signal to form said word construct.

16. A computer readable medium consisting of media selected from among ROM, RAM, magnetic disk storage media, optical storage media, and flash memory devices, said media containing executable instruction, which, when executed in a processing system, cause said processing system to perform a text entry method comprising the steps of:

receiving a plurality of input signals from a reduced keyboard by a user, each input signal corresponding to a key within a user input interface and further representing language phonetic equivalents of at least one language character comprising a language character intended by said user to form a word construct in a language, wherein at least one series of said plurality of input signals comprises a consonant plus a matra plus a sign cluster combining to represent a conjunct, wherein said plurality of input signals comprise all of;

a first input signal representing a Latin phonetic equivalent of at least one character; and

at least one additional input signal in a particular position, following a switching signal indicating a mode change, wherein said at least one additional input signal is translated into a Hindu, Bengali, or Tamil character, forming a specific language character intended for a particular position; and

using said first input signal to generate at least a first proposed language character;

using said at least one additional input signal to generate at least a second proposed language character;

forming at least one list of possible word constructs using said first proposed language character and said second proposed language character;

filtering possible word constructs not containing said at least one additional input signal in the intended position from said list of possible word constructs, forming a filtered list of word constructs; and

displaying said filtered list of word constructs for said user on a display screen.

17. A computer readable medium consisting of media selected from among ROM, RAM, magnetic disk storage media, optical storage media, and flash memory devices, said media containing executable instructions which, when executed in a processing system, cause said processing system to perform a text entry method comprising the steps of:

receiving a plurality of input signals from a keyboard by a user, each input signal corresponding to a key within a user input interface and further representing language phonetic equivalents of at least one language character comprising a language character intended by said user to form a word construct in a language, wherein said plurality of input signals are entered on a standard telephone input keypad, wherein all eleven vowel characters, thirty-five consonant characters, and three special sign characters needed to spell Bengali words are contained across keys 2 through 0 of said standard telephone input keypad, wherein said input signals comprise all of;

a first input signal representing a Latin phonetic equivalent of at least one character; and

at least one additional input signal, following a switching signal indicating a mode change, wherein said at least one additional input signal is translated into a Hindu, Bengali, or Tamil character, forming a specific language character intended for a particular position; and

using said each input signal to generate at least one proposed language character, wherein the first input signal and the at least one additional input signal forms a list of possible word constructs;

filtering possible word constructs not containing said proposed language characters in the intended positions from said list of possible word constructs, forming a filtered list of word constructs containing potential word constructs;

displaying successively said at least one proposed language character for said user on a display screen; and

facilitating successive selection by said user of potential word constructs.