US6470306B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Automated translation of annotated text based on the determination of locations for inserting annotation tokens and linked ending, end-of-sentence or language tokens
Assignee: CANTENA SERVICE AGENT Corp
•
Filing: 1997-06-09
•
Granted: 2002-10-22
•
Expires: 2017-06-09
Abstract
Abstract A system and method for translating an annotated source document in a first natural language to a target document in a second natural language having corresponding annotations, includes computer storage, a computer receiving module for receiving input textual information in a first language and for storing the input textual information in the computer storage, the input textual information including annotations and a translation engine for creating a first token string including first language tokens, annotation tokens that apply to the first language tokens, and ending tokens. Prior to translation, the annotation tokens are removed from the first token string and a plurality of annotation records are created for the first language tokens, each annotation record linking one of the first language tokens to each of the annotation tokens that apply to the first language token. After the first language tokens have been translated into second language tokens, the annotation tokens are inserted at appropriate locations in the second token string. These locations are determined using the end-of-sentence tokens and the annotation records, and a target document is produced in the second language using the second token string and annotations, thus recreating in the target document, the annotations of the source document.
Inventors
- 👤 Lewis G. Pringle
- 👤 Robert W. Swerdlow
- 👤 Alec Wysoker
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 10:32
CPC Classifications
G; G06; G06F; G06F40/00; G06F40/10; G06F40/166; G06F40/169; G06F40/40; G06F40/58
Claims Analysis
Total Claims: 39
Claim 1
Independent Claim
210 words
A method for translating an annotated source document in a first language to an annotated source document in a second language, comprising: receiving a source document in a first language, the source document comprising a plurality of sentences, the sentences having text and annotations; creating a first token string using the source document, the first token string comprising a plurality of first language tokens and a plurality of annotation tokens; inserting a plurality of end-of-sentence tokens into the token string at locations corresponding to sentence endings in the source document; removing the plurality of annotation tokens from the token string; storing the removed annotation tokens in a storage module and linking the annotation tokens to the end-of-sentence tokens in the storage module; translating the plurality of first language tokens and creating a second token string comprising a plurality of second language tokens and the end-of-sentence tokens; determining locations for inserting each of the plurality of annotation tokens in the second token string using the end-of-sentence tokens; and inserting the plurality of annotation tokens into the second token string at the locations; and producing a translated version of the source document in the second language using the plurality of second language tokens and the annotation tokens in the second token string.
Claim 2
Independent Claim
210 words
A method for translating an annotated source document in a first language to an annotated source document in a second language, comprising: receiving a source document in a first language, the source document comprising a plurality of sentences, the sentences having text and annotations; creating a first token string using the source document, the first token string comprising a plurality of first language tokens and a plurality of annotation tokens; inserting a plurality of end-of-sentence tokens into the token string at locations corresponding to sentence endings in the source document; removing the plurality of annotation tokens from the token string; storing the removed annotation tokens in a storage module and linking the annotation tokens to the end-of-sentence tokens in the storage module; translating the plurality of first language tokens and creating a second token string comprising a plurality of second language tokens and the end-of-sentence tokens; determining locations for inserting each of the plurality of annotation tokens in the second token string using the end-of-sentence tokens; and inserting the plurality of annotation tokens into the second token string at the locations; and producing a translated version of the source document in the second language using the plurality of second language tokens and the annotation tokens in the second token string.
Claim 3
Dependent Claim
28 words
The method of claim 1 , wherein the annotation tokens represent one or more of the following: part-of-speech settings, HTML markup, SGML markup, RTF markup, and NROFF markup.
Claim 4
Dependent Claim
22 words
The method of claim 1 , wherein the first language tokens represent English text and the second language tokens represent Japanese text.
Claim 5
Dependent Claim
22 words
The method of claim 1 , wherein the first language tokens represent English text and the second language tokens represent Spanish text.
Claim 6
Dependent Claim
62 words
The method of claim 1 , further comprising: storing in a sentence database corresponding to the order of appearance in a sentence in the source document, the plurality of first language tokens, the annotations tokens and the end-of-sentence tokens; and using the tokens stored in the sentence database to determine the locations for inserting the annotation tokens into the second token string.
Claim 7
Dependent Claim
56 words
The method of claim 1 , further comprising: determining which of the first language tokens in the first language token string are undefined first language tokens; storing the undefined first language tokens in the storage module; and providing a list of undefined first language tokens to the viewer of the translated version of the source document.
Claim 8
Dependent Claim
44 words
The method of claim 1 , further comprising: storing the first language tokens in the storage module; storing the second language tokens in the storage module; and using the end-of-sentence tokens to link the stored first language tokens with the stored second language tokens.
Claim 9
Dependent Claim
29 words
The method of claim 1 , further comprising: determining the end of the source document, creating an end-of-file token, and inserting the end-of-file token in the first token string.
Claim 10
Dependent Claim
18 words
The method of claim 1 , wherein the annotation tokens comprise HTML tokens representing HTML character entity references.
Claim 11
Dependent Claim
41 words
The method of claim 9 , the step of creating a first token string further comprising: determining a substitute character for a character entity reference in at least one HTML token; and replacing the character entity reference with the substitute character.
Claim 12
Dependent Claim
34 words
The method of claim 9 , further comprising: determining which of the HTML tokens comprise non-translatable HTML markup characters; and removing from the first token string, the HTML tokens comprising non-translatable HTML markup characters.
Claim 13
Dependent Claim
41 words
The method of claim 9 , further comprising: storing the plurality of HTML tokens in a document state database; and creating a vector linking each of the plurality of HTML tokens with each of the plurality of the first language tokens.
Claim 14
Dependent Claim
20 words
The method of claim 12 , further comprising: inserting the HTML tokens into the second token string using the vector.
Claim 15
Dependent Claim
21 words
The method of claim 9 , wherein the text further comprises text that alternately replaces an image in the source document.
Claim 16
Dependent Claim
47 words
The method of claim 1 , further comprising: determining a location in the source document between a beginning of the source document and an ending of the source document; translating the first token string from the location; and producing a partially translated version of the source document.
Claim 17
Independent Claim
181 words
A system for translating an annotated source document in a first language, to an annotated target document in a second language while preserving annotations from the source document to the target document, comprising: a receiving module for receiving a source document in a first language, the source document comprising text and annotations; a processing module for creating a first token string comprising a plurality of first language tokens, a plurality of annotation tokens, and a plurality of ending tokens corresponding to discontinuities in the text of the source document; and a translation engine for removing the plurality of annotation tokens from the first token string, translating the plurality of first language tokens to a plurality of second language tokens in a second token string; using the plurality of ending tokens to insert the annotation tokens into the second token string, and creating a target document comprising a translated version of the source document using the second token string; and a storage module comprising an annotation database for storing the annotation tokens, wherein the annotation tokens are linked to the ending tokens.
Claim 18
Independent Claim
181 words
A system for translating an annotated source document in a first language, to an annotated target document in a second language while preserving annotations from the source document to the target document, comprising: a receiving module for receiving a source document in a first language, the source document comprising text and annotations; a processing module for creating a first token string comprising a plurality of first language tokens, a plurality of annotation tokens, and a plurality of ending tokens corresponding to discontinuities in the text of the source document; and a translation engine for removing the plurality of annotation tokens from the first token string, translating the plurality of first language tokens to a plurality of second language tokens in a second token string; using the plurality of ending tokens to insert the annotation tokens into the second token string, and creating a target document comprising a translated version of the source document using the second token string; and a storage module comprising an annotation database for storing the annotation tokens, wherein the annotation tokens are linked to the ending tokens.
Claim 19
Dependent Claim
75 words
The system of claim 16 , the storage module further comprising: a dictionary source database for storing the first language tokens and the second language tokens, wherein the ending tokens provide links between the first language tokens and the second language tokens in the dictionary source database, and an undefined words database for storing undefined first language tokens, wherein the ending tokens provide links to the undefined first language tokens in the undefined words database.
Claim 20
Dependent Claim
48 words
The system of claim 17 , wherein the translation engine accesses the markup database and compares the second token string with HTML markup linked to the first language tokens in the markup database to determine locations in the second token string where the HTML tokens should be inserted.
Claim 21
Dependent Claim
20 words
The system of claim 16 , wherein the processing module creates HTML tokens representing HTML markup in the source document.
Claim 22
Dependent Claim
32 words
The system of claim 19 , wherein the annotation database comprises a markup database for linking each first language token in the first token string to the HTML tokens that apply thereto.
Claim 23
Independent Claim
188 words
A method for translating an HTML annotated source document in a first language to a target document in a second language having corresponding HTML markup, comprising: receiving a source document in a first language, the source document comprising a plurality of sentences, the sentences having text and HTML markup; creating a first token string using the source document, the first token string comprising a plurality of first language tokens and a plurality of HTML tokens that apply to the first language tokens; removing the HTML tokens from the first token string; creating a plurality of annotation records for the first language tokens, each annotation record linking one of the first language tokens to each of the HTML tokens that apply to the first language token; storing the annotation records in a document state database; translating the plurality of first language tokens and creating a second token string comprising a plurality of second language tokens; and determining at which locations in the second token string the HTML tokens should be inserted using the annotation records; and producing a target document in the second language using the second token string.
Claim 24
Independent Claim
188 words
A method for translating an HTML annotated source document in a first language to a target document in a second language having corresponding HTML markup, comprising: receiving a source document in a first language, the source document comprising a plurality of sentences, the sentences having text and HTML markup; creating a first token string using the source document, the first token string comprising a plurality of first language tokens and a plurality of HTML tokens that apply to the first language tokens; removing the HTML tokens from the first token string; creating a plurality of annotation records for the first language tokens, each annotation record linking one of the first language tokens to each of the HTML tokens that apply to the first language token; storing the annotation records in a document state database; translating the plurality of first language tokens and creating a second token string comprising a plurality of second language tokens; and determining at which locations in the second token string the HTML tokens should be inserted using the annotation records; and producing a target document in the second language using the second token string.
Claim 25
Dependent Claim
32 words
The method of claim 21 , further comprising: determining whether any of the HTML tokens in the first token string comprise character entity references; and substituting characters for the character entity references.
Claim 26
Dependent Claim
39 words
The method of claim 21 , further comprising: determining whether any of the HTML tokens should not be preserved in the second token string; and deleting from the first token string, the HTML tokens that should not be preserved.
Claim 27
Dependent Claim
57 words
The method of claim 21 , further comprising: determining whether any of the tokens in the first token string should not be translated; removing the tokens that should not be translated from the first token string; storing the removed tokens; and inserting marker tokens into the first token string in the locations where the tokens were removed.
Claim 28
Dependent Claim
52 words
The method of claim 21 , further comprising: determining whether the HTML tokens represent a discontinuity in the source text; inserting into the first token string, ending tokens representing the discontinuity; and storing in a database indexed by the ending token, the tokens in the first token string up to the discontinuity.
Claim 29
Dependent Claim
22 words
The system of claim 21 , wherein the first language tokens represent English text and the second language tokens represent Japanese text.
Claim 30
Dependent Claim
22 words
The system of claim 21 , wherein the first language tokens represent English text and the second language tokens represent Spanish text.
Claim 31
Dependent Claim
230 words
An automated natural language translation system, comprising: computer storage means; means for receiving input textual information in a first language and for storing the input textual information in the computer storage means, the input textual information including annotations; and a translation engine for accessing the computer storage means and for translating the input textual information in the first language into output textual information in a second language, the translation engine comprising: (i) a tokenizer for creating a first token string using the source document, the first token string comprising a plurality of first language tokens and a plurality of annotation tokens that apply to the first language tokens, (ii) a filter for removing a subset of the annotation tokens from the first token string, (iii) a document state recorder for creating a plurality of annotation records for the first language tokens, each annotation record linking one of the first language tokens to each of the annotation tokens that apply to the first language token, (iv) a translator for translating the plurality of first language tokens and creating a second token string comprising a plurality of second language tokens, (v) an annotator for determining at which locations in the second token string the annotation tokens should be inserted using the annotation records, and (vi) a terminal consumer for producing a target document in the second language using the second token string.
Claim 32
Dependent Claim
230 words
An automated natural language translation system, comprising: computer storage means; means for receiving input textual information in a first language and for storing the input textual information in the computer storage means, the input textual information including annotations; and a translation engine for accessing the computer storage means and for translating the input textual information in the first language into output textual information in a second language, the translation engine comprising: (i) a tokenizer for creating a first token string using the source document, the first token string comprising a plurality of first language tokens and a plurality of annotation tokens that apply to the first language tokens, (ii) a filter for removing a subset of the annotation tokens from the first token string, (iii) a document state recorder for creating a plurality of annotation records for the first language tokens, each annotation record linking one of the first language tokens to each of the annotation tokens that apply to the first language token, (iv) a translator for translating the plurality of first language tokens and creating a second token string comprising a plurality of second language tokens, (v) an annotator for determining at which locations in the second token string the annotation tokens should be inserted using the annotation records, and (vi) a terminal consumer for producing a target document in the second language using the second token string.
Claim 33
Dependent Claim
19 words
The system of claim 28 , wherein the annotations comprise HTML markup, SGML markup, RTF markup or NROFF markup.
Claim 34
Dependent Claim
22 words
The system of claim 28 , wherein the first language tokens represent English text and the second language tokens represent Japanese text.
Claim 35
Dependent Claim
22 words
The system of claim 28 , wherein the first language tokens represent English text and the second language tokens represent Spanish text.
Claim 36
Dependent Claim
36 words
The system of claim 28 , wherein the translation engine further comprises: means for receiving a user input; and an alternate translator for processing an input from a user and providing translation options to the user.
Claim 37
Dependent Claim
34 words
The system of claim 32 , further comprising an alternate text producer for receiving a source document and an input from a user and transmitting the source document and the input to the tokenizer.
Claim 38
Dependent Claim
18 words
The system of claim 33 , wherein the input from a user comprises edits to the source document.
Claim 39
Dependent Claim
28 words
The system of claim 34 , wherein the plurality of first language tokens created by the tokenizer comprise first language tokens representing the edits to the source document.
References Cited (Backward Citations)
US4791587A
Cited
US4964044A
Cited
US4833611A
Cited
US4805100A
Cited
US4916614A
Cited
US4980829A
Cited
US4958285A
Cited
US4868750A
Cited
EP0357344A2
Cited
US5214583A
Cited
US5136503A
Cited
US5060155A
Cited
US4984178A
Cited
US5029085A
Cited
US5243520A
Cited
US5497319A
Cited
US5321607A
Cited
US5587902A
Cited
US5528491A
Cited
US5448474A
Cited
US5987402A
Cited
US5890103A
Cited
WO1997018516A1
Cited
US5930746A
Cited
US5848386A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2005-10-21 | 2006-04-22 | 2006-10-22 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2009-10-21 | 2010-04-22 | 2010-10-22 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2013-10-21 | 2014-04-22 | 2014-10-22 | $7700.00 | Window Closed / Lapsed |