Translation system using a translation database, translation using a translation database, method using a translation database, and program for translation using a translation database

• US 7,797,150 B2
• Filed: 09/12/2005
• Issued: 09/14/2010
• Est. Priority Date: 03/09/2005
• Status: Expired due to Fees

• Current Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Original Assignee: Fuji Xerox Company Limited (Xerox Holdings Corp.)
• Inventors: Yoshimura, Hiroki, Kimura, Shunichi
• Primary Examiner(s): Opsasnick; Michael N
• Application Number:
  US11/223,009
• Publication Number:
  US 20060206305A1
• Time in Patent Office: 1,828 Days
• Field of Search: None
• US Class Current: 704/8
• CPC Class Codes: G06F 40/45 Example-based machine translation; Alignment
• 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

A translation system comprises an image reading unit that optically reads an image of a manuscript and generates image data; an inputting unit that inputs a translation target language; a character recognizing unit that generates an original text by performing a character recognition process on the image data generated by the image reading unit; a translation text database in which are associated and stored translation texts, language identifiers which specify the languages in which the translation texts are written, and document identifiers which specify the contents of the translation texts; an extracting unit that extracts the document identifier which specifies the content of the original text from the original text; a searching unit that searches the translation text database for a translation text associated with a document identifier identical to the document identifier extracted from the original text by the extracting unit and a language identifier identical to the language identifier which specifies the translation target language input by the inputting unit; and an outputting unit that outputs the translation text searched by the searching unit.

Claims

1. A translation system comprising:

an image reading unit that optically reads an image of a manuscript and generates image data;

an inputting unit that inputs a translation target language;

a character recognizing unit that generates an original text by performing a character recognition process on the image data generated by the image reading unit;

a translation database that stores translation phrases in a plurality of languages, language identifiers that specify the language in which each individual translation phrase is written, and document identifiers that identify a content of the translation phrase regardless of the language in which the translation phrase is written;

an extracting unit that extracts the document identifier which specifies the content of the original text from the original text;

a searching unit that searches the translation database for a translation text associated with a document identifier identical to the document identifier extracted from the original text by the extracting unit and a language identifier identical to the language identifier which specify the translation target language input by the inputting unit; and

an outputting unit that outputs the translation text searched by the searching unit

wherein if the translation text associated with the document identifier is not found in the translation database, the original text is translated to generate a translation text for the original text and the language identifier, the document identifier, and the generated translation text are automatically stored in the translation database.

2. The translation system according to claim 1, further comprising a notifying unit that notifies that the searching unit can find no translation phrases that satisfy the condition in the translation database, if none can be found.

3. The translation system according to claim 1, further comprising a translating unit that generates a translation phrase by translating the original text into the translation target language if no translation texts satisfying the condition can be found in the translation database.

4. A translation system comprising:

an image reading unit that optically reads an image of a manuscript and generates image data;

an inputting unit that inputs a translation target language;

a character recognizing unit that generates an original text by performing a character recognition process on the image data generated by the image reading unit;

a translation database that stores translation phrases in a plurality of languages, language identifiers that specify the language in which each individual translation phrase is written, and intermediate language texts generated from the translation phrases;

an intermediate language generating unit that translates the original text into an intermediate language and generates an intermediate language text;

a proximity calculating unit that calculates the proximity between the intermediate language text obtained by the intermediate language generating unit and the intermediate language text stored in the translation database; and

a searching unit that searches the translation database for a translation text whose proximity calculated by the proximity calculating unit satisfies a prescribed condition and which is associated with a language identifier identical to the language identifier which specify the translation target language input by the inputting unit,

wherein if the translation text associated with the document identifier is not found in the translation database, the original text is translated to generate a translation text for the original text and the language identifier, the document identifier, and the generated translation text are automatically stored in the translation database.

5. The translation system according to claim 4, wherein the prescribed condition is the condition that a minimum value of the proximities calculated by the proximity calculating unit be not greater than a threshold value.

6. The translation system according to claim 4, further comprising a notifying unit that notifies that the searching unit can find no translation phrases that satisfy the condition in the translation database, if none can be found.

7. The translation system according to claim 4, further comprising a translating unit that generates a translation phrase by translating the original text into the translation target language if no translation texts satisfying the condition can be found in the translation database.

8. A translation method for translating a manuscript according to a translation database that stores translation phrases in a plurality of languages, language identifiers that specify the language in which each individual translation phrase is written, and document identifiers that identify a content of the translation phrase regardless of the language in which the translation phrase is written, the method comprising:

optically reading an image of a manuscript and generates image data;

inputting a translation target language;

generating an original text by performing a character recognition process on the image data;

extracting from the original text a document identifier that identifies a content of the original text regardless of the language in which the original text is written;

searching the translation database for a translation text associated with a document identifier identical to the document identifier extracted from the original text by the extracting step and a language identifier identical to the language identifier which specifies the input translation target language; and

outputting the searched translation text,

wherein if the translation text associated with the document identifier is not found in the translation database, the original text is translated to generate a translation text for the original text and the language identifier, a document identifier, and the generated translation text are automatically stored in the translation database.

9. A translation method for translating a manuscript according to a translation database that stores translation phrases in a plurality of languages, language identifiers that specify the language in which each individual translation phrase is written, and intermediate language texts generated from the translation phrases, the method including:

optically reading an image of a manuscript and generates image data;

inputting a translation target language;

generating an original text by performing a character recognition process on the image data;

generating an intermediate language text by translating the original text into an intermediate language;

calculating the proximity between the intermediate language text and the intermediate language text; and

searching the translation database for a translation text whose proximity satisfies a prescribed condition and which is associated with a language identifier identical to the language identifier which specifies the input translation target language,

wherein if the translation text associated with the document identifier is not found in the translation database, the original text is translated to generate a translation text for the original text and the language identifier, the document identifier, and the generated translation text are automatically stored in the translation database.

10. A storage medium readable by a computer, the storage medium storing a program of instructions executable by a computer device, the computer device including a translation database that stores translation phrases in a plurality of languages, language identifiers that specify the language in which each individual translation phrase is written, and document identifiers that identify a content of the translation phrase regardless of the language in which the translation phrase is written, the program comprising:

an image reading step that optically reads an image of a manuscript and generates image data;

an inputting step that inputs a translation target language;

a character recognizing step that generates an original text by performing a character recognition process on the image data;

an extraction step that extracts from the original text a document identifier that identifies a content of the original text regardless of the language in which the original text is written;

a searching step that searches the translation database for a translation text associated with a document identifier identical to the document identifier extracted from the original text by the extracting step and a language identifier identical to the language identifier which specifies the input translation target language; and

an outputting step that outputs the translation text searched for in the searching step,

wherein if the translation text associated with the document identifier is not found in the translation database, the original text is translated to generate a translation text for the original text and the language identifier, the document identifier, and the generated translation text are automatically stored in the translation database.

11. A storage medium readable by a computer, the storage medium storing a program of instructions executable by a computer device, the computer device including a translation database that stores translation phrases in a plurality of languages, language identifiers that specify the language in which each individual translation phrase is written, and intermediate language texts generated from the translation phrases, the program comprising:

an image reading step that optically reads an image of a manuscript and generates image data;

an inputting step that inputs a translation target language;

a character recognizing step that generates an original text by performing a character recognition process on the image data;

an intermediate language generating step that translates the original text into an intermediate language and generates an intermediate language text;

a proximity calculating step that calculates the proximity between the intermediate language text and the intermediate language text stored in the translation database; and

a searching step that searches the translation database for a translation text whose proximity satisfies a prescribed condition and which is associated with a language identifier identical to the language identifier which specifies the input translation target language,

wherein if the translation text associated with the document identifier is not found in the translation database, the original text is translated to generate a translation text for the original text and the language identifier, a document identifier, and the generated translation text are automatically stored in the translation database.