System and method for distributing multilingual documents

• US 7,493,250 B2
• Filed: 12/18/2000
• Issued: 02/17/2009
• Est. Priority Date: 12/18/2000
• Status: Expired due to Fees

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Luckman, Colin, Kaplan, Ronald M., Petrie, Glen W., Hecht, David L.
• Primary Examiner(s): Wozniak; James S
• Application Number:
  US09/738,291
• Publication Number:
  US 20020077805A1
• Time in Patent Office: 2,983 Days
• Field of Search: 235/494, 235/487, 235/454, 382/182, 382/181, 704 2- 3, 704 7- 8, 358/3.28, 282/232, 715/530
• US Class Current: 704/3
• CPC Class Codes: G06F 40/42 Data-driven 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

A method and apparatus is disclosed for generating and distributing multilingual documents. The multilingual documents are comprised of primary information consisting of human-readable text and secondary information consisting of machine-readable data such that a translation of the text is accomplished by converting the human-readable text into a second language through the use of the decoded machine-readable data. The machine-readable data is comprised of a code that describes a set of editing operations that can be applied to the human-readable text to convert it into at least a second language. In a preferred embodiment, the machine-readable data is embedded in the image using an unobtrusive code on the document such as Xerox DATAGLYPH codes.

Claims

1. A method for converting a document from a first language into a second language comprising the steps of:

receiving, via an electronic device, image data indicating a document, wherein said document, when rendered, comprises human-readable text written in a first language;

said image data including language translation data encoded, via an encoding module, in binary machine-readable code embedded in said image data such that when said document is rendered, the encoded language translation data is both rendered on the document and not human-readable;

receiving selection data indicating a selected foreign language for translation of said human-readable text written in the first language;

producing, via a decoding module, a human-readable translation of said document in said selected foreign language using the language translation data encoded in said machine-readable code;

wherein the encoded language translation data represents a correction code C that describes a set of editing functions E that are applied to the text, which is human readable text P, of the document having the first language, to convert the text P from the first language into the second language, wherein for each page of the text P in the first language, there is an accurate translation ATL into the second language L;

wherein a processing routine RL, is applied to each page of the text P in the first language to produce the translation of the text P into the second language L, the quality of the translation RL(P) being on a continuum from very good to very bad;

wherein the code C is computed such that;

ATL=E(C, RL(P)),

and the code C is transmitted as glyphs on the page containing the text P; and

wherein multilingual encoding and decoding modules reconstruct the accurate translation ATL by optical character recognition processing of the text P, applying the processing routine RL to the translation result, and then correcting the translation result according to instructions of the code C;

using a Machine Translation MT capability to produce a translation by optical character recognition processing of the text of the document in the first language; and

after applying the Machine Translation MT to the text P, the steps of C and E performing additional processing to improve readability of the text P, wherein the code C contains the operations that the editing functions E perform to produce the ATL, including performing disambiguation by framing a series of questions to a person fluent with the first language, and using by the translation software answers to the questions to make choices of word sense and sentence patterns in the second language, the code C recording the answers to the series of questions so the fluent person'"'"'s knowledge is available for guidance when a translation is undertaken.

2. The method according to claim 1 wherein in addition to providing the correction code C, further including secondary information that also encodes information that describes at least one of,

(i) an encoding scheme, (ii) a compression algorithm, (iii) settings such as one of a font identifier, error correction data, or codes for characters, (iv) datasets, and (v) hints that are used to translate the text of the document.