Language translation system and method

• US 5,765,131 A
• Filed: 01/24/1995
• Issued: 06/09/1998
• Est. Priority Date: 10/03/1986
• Status: Expired due to Fees

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Stentiford, Frederick W., Steer, Martin G.
• Primary Examiner(s): Zele, Krista
• Assistant Examiner(s): Wolinsky, Scott
• Application Number:
  US08/377,599
• Time in Patent Office: 1,232 Days
• Field of Search: 379/67, 379/84, 379/88, 379/89, 379/37, 364/900, 364/513.5, 364/419, 364/419.03, 340/711, 395/2.66, 704/257, 704/2, 704/3, 704/4, 704/8, 704/277, 704/235, 704/260
• US Class Current: 704/277
• CPC Class Codes: G06F 40/45 Example-based machine translation; Alignment

Abstract

A language translation system for translating phrases from a first language into a second language comprises a store holding a collection of phrases in the second language. Phrases input in the first language are each characterized on the basis of one or more keywords, and the corresponding phrase in the second language is output. Such a phrasebook approach enables what is effectively rapid and accurate translation, even from speech. Since the phrases in the second language are prepared in advance and held in store, there need be no problems of poor translation or ungrammatical construction. The output may be in text, or, using speech synthesis, in voiced form. With appropriate choice of keywords it is possible to characterize a large number of relatively long and complex phrases with just a few keywords.

Claims

1. A language translation system comprising:

input means for accepting an input phrase consisting of a plurality of words;

at least one store holding a first collection of phrases in a first language and a second collection of phrases in a second language which correspond to the first collection of phrases;

characterization means connected to the input means and to the at least one store for determining which phrase of at least one of said first and second collections is associated with the input phrase, the characterization means including means coupled to the at least one store for identifying among the words of the input phrase, the presence of at least one keyword of a predetermined set of keywords, the characterization means being operative in response to the identification of a given keyword or given combination of keywords to select a phrase from said at least one of said first and second collections of phrases; and

output means responsive to the characterization means for outputting the selected phrase as a confirmatory audio message;

the system thereafter producing a further output in dependence upon the identity of the selected phrase.

2. A system as in claim 1 wherein the number of members in the set of keywords is smaller than the number of phrases in the collections.

3. A system as claimed in claim 1 wherein the characterization means comprises a speech recognizer.

4. A system as claimed in claim 1, wherein said input means is capable of accepting spoken inputs, and said output means provides voiced outputs.

5. A system as claimed in claim 1 wherein means are provided to enable portions of said input phrase to be passed to said output means for outputting as part of the selected phrase.

6. A system as claimed in claim 1 further comprising a keyboard for providing an input message to said input means, and means to provide a text output.

7. A system as claimed in claim 1 further comprising a further collection of phrases in each of a plurality of second languages.

8. A system according to claim 1, in which each phrase of said collections contain a unique keyword, keyword-part or combination of keywords or keyword-parts.

9. A system according to claim 1 in which the characterization means is operable in the case that more than one keyword is recognized in the input phrase to make use of their relative positions within the input phrase for the purpose of distinguishing between phrases of the collections.

10. A system as claimed in claim 1, in which the characterization means applies a first set of keywords to determine to which phrase or group of phrases, if any, from said collections of phrases the input phrase corresponds, and, in the case that the input phrase is found to correspond to an undetermined one of a group of phrases, the characterization means applies a second set of keywords to determine to which one of the group of phrases the input phrase corresponds.

11. A method of translating multi-word phrases, the method comprising:

accepting an input phrase consisting of a plurality of words;

identifying among the words of the input phrase the presence of at least one keyword of a predetermined set of keywords;

in response to the identification of a given keyword or given combination of keywords in a second language, selecting a phrase from a first collection of phrases in a first language and a second collection of phrases, which correspond to the first collection of phrases;

outputting the selected phrase as a confirmatory audio message; and

thereafter producing a further output in dependence upon the identity of the selected phrase.