Translation system and method in which words are translated by a specialized dictionary and then a general dictionary

• US 6,085,162 A
• Filed: 10/18/1996
• Issued: 07/04/2000
• Est. Priority Date: 10/18/1996
• Status: Expired due to Fees

• Current Assignee: Gedanken Corporation inc.
• Original Assignee: Gedanken Corporation inc.
• Inventors: Cherny, Julius
• Primary Examiner(s): Voeltz, Emanual Todd
• Assistant Examiner(s): SOFOCLEOUS, MICHAEL D
• Application Number:
  US08/733,808
• Time in Patent Office: 1,355 Days
• Field of Search: 704/277, 704/2, 704/10, 704/250, 704/256
• US Class Current: 704/277
• CPC Class Codes: G06F 40/242 Dictionaries
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 40/44 Statistical methods, e.g. probability models
• CPC Class Codes: G06F 40/55 Rule-based translation
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation

Abstract

Methods and apparatus for performing translation between different language are provided. The present invention includes a translation system that performs translation having increased accuracy by providing a three-dimensional topical dual-language database. The topical database includes a set of source-to-target language translations for each topic that the database is being used for. In one embodiment, a user first selects the topic of conversation, then words spoken into a telephone are translated and produced as synthesized voice signals from another telephone so that a near real-time conversation may be had between two people speaking different languages. An additional feature of the present invention is the addition of a computer terminal that displays the input and output phrases so that either user may edit the input phrases, or indicate that the translation was ambiguous and request a rephrasing of the material.

Claims

1. A method for translating information from a source language to a target language based on the selection of a topic by a user, said method comprising:

selection of said topic by said user;

inputting said information in said source language to produce input words;

automatically processing said information based on said topic selection, said processing comprising;

classifying said input words;

applying said input words against a topical dictionary comprising at least one layer of bi-directional translations formed via a frequency-of-association comparison of a word file in said source language and a word file in said target language based on said topic selected by said user, said source and target language word files each being formed from a plurality of materials specific to said selected topic, said application of said input words comprising producing one or more translations of said input words into said target language and outputting one or more untranslated words;

forming sentences from said input words based on a set of syntax rules for said source language;

applying a conventional dual-language dictionary to said untranslated words to provide at least one translation of said untranslated words into said target language, said at least one translation being provided only if said topical dictionary failed to provide one or more translations for said untranslated words;

processing said input words, said at least one translation, and said one or more translations against a set of syntax rules for said target language to produce a translated message; and

outputting said translated message.

2. The method of claim 1, wherein said step of outputting said translated message includes the step of:

displaying the translated message on a computer screen.

3. The method of claim 2 further comprising the step of:

enabling the recipient of the translated message to indicate that one or more words of the translated message are unclear.

4. The method of claim 1, wherein said step of inputting comprises the steps of:

receiving oral voice signals from a telephone; and

processing said voice signals using voice recognition circuitry to produce electronic signals representative of said voice signals.

5. The method of claim 1, wherein said step of inputting comprises the steps of:

receiving signals from a scanner; and

processing said scanner signals using optical character recognition circuitry to convert said scanned signals into textual information.

6. The method of claim 1, wherein said step of inputting comprises the step of:

receiving textual information via a computer keyboard.

7. The method of claim 1, wherein said step of classifying comprises the step of:

determining the grammatical function of said input words.

8. The method of claim 1, wherein said step of classifying comprises the step of;

determining the grammatical form of said input words.

9. The method of claim 1, wherein said step of classifying comprises the step of:

determining the denotation of said input words.

10. The method of claim 1, wherein said step of classifying comprises the steps of:

determining the grammatical function of said input words;

determining the grammatical form of said input words; and

determining the denotation of said input words.

11. The method of claim 1, wherein said step of classifying includes the step of:

applying said input words against a Markov model that associates each input word against the input words surrounding each input word to reduce homonym occurrences.

12. The method of claim 1, wherein said step of outputting said translated message includes the steps of:

synthesizing said translated message into voice signals; and

providing said voice signals to a telephone system so that they may be output by a telephone.

13. The method of claim 1, wherein said step of outputting comprises the step of:

printing out the translated message.

14. A system for providing translation between a source language and a target language based on the selection of a topic by a user, said system comprising:

circuitry for determining said selection of said topic by said user;

input circuitry for receiving input messages in said source language;

output circuitry for automatically producing translated messages in said target language based on said selection of said topic by said user;

memory including partitions having a topical dictionary comprising at least one layer of bi-directional translations formed via a frequency-of-association comparison of a word file in said source language and a word file in said target language, said source and target language word files each being formed from a plurality of materials specific to said selected topic, a dual-language dictionary, syntax rules for said source language and for said target language, and grammatical rules for said source language and for said target language; and

processing circuitry coupled to said input circuitry, to said output circuitry and to said memory, said processing circuitry processing said input messages from said source language into said target language to produced a translated message, at least by applying said input message against said topical dictionary to produce one or more translations of said input messages and one or more untranslated words, and by applying a conventional dual-language dictionary to said untranslated words to produce at least one translation of said untranslated words, said translated message being produced at least from said one or more translations and said at least one translation.

15. The system for translating messages of claim 14, wherein said input circuitry comprises:

a computer terminal including a video screen and keyboard.

16. The system for translating messages of claim 15, wherein said video screen displays said input messages and said computer terminal further comprises means for revising said input message.

17. The system for translating messages of claim 14, wherein said output circuitry comprises:

a computer terminal including a video screen and keyboard.

18. The system for translating messages of claim 17, wherein said computer terminal further comprises:

means for indicating to said user that one or more words of said translated message are unclear.

19. The system for translating messages of claim 14, wherein said input circuitry comprises:

a telephone; and

voice recognition circuitry.

20. The system for translating messages of claim 14, wherein said input circuitry comprises:

a scanner; and

optical character recognition circuitry.

21. The system for translating messages of claim 14, wherein said output circuitry comprises:

a telephone; and

voice synthesis circuitry.

22. The system for translating messages of claim 14, wherein said output circuitry comprises:

a printer.