General dictionary for all languages

1. An electronic apparatus configured to translate words into a target language, the electronic apparatus comprising:

an electronic display;

an electronic processor; and

instructions to cause the electronic apparatus to:

acquire an image of text;

detect a selection of a word or word combination in the image to be translated;

perform optical character recognition (OCR) on the selected word or word combination using a character alphabet of a plurality of languages;

generate a set of recognition variants for each word of the selected word or word combination;

transmit each set of recognition variants to a set of language specific processors;

eliminate language inappropriate variants from the set of recognition variants for each language, wherein the language inappropriate variants are the recognition variants which do not contain characters or symbols of the language;

match each of remaining variants to a source language, wherein the remaining variants are the recognition variants minus the language inappropriate variants;

confirm that at least one of the remaining variants is in at least one language specific word list;

translate a confirmed word variant using a translation dictionary; and

provide a translation of the confirmed word variant.

2. The electronic apparatus of claim 1 , wherein the instructions further cause the apparatus to:

display proposed words from the language specific word list based on an unconfirmed recognition variant;

detect selection of a proposed word or correction of a variant; and

provide a translation of the proposed word or correction of the variant.

3. The electronic apparatus of claim 1 , wherein said matching of each of the remaining variants to a source language includes:

matching language specific characters to the source language.

4. The electronic apparatus of claim 1 , wherein said eliminating language inappropriate variants from the set of recognition variants includes:

eliminating an inappropriate variant based on detection of a language inconsistent character.

5. The electronic apparatus of claim 1 , wherein the instructions further cause the electronic apparatus to:

prior to detecting the selection of a word or word combination in the image, perform OCR on a portion of the image of text to generate recognized words;

locate recognized words in monolingual dictionaries;

generate hypotheses about one or more source languages of the text;

rank hypotheses based on a frequency of each language's use in the text; and

limit, based on the ranked hypotheses, the character alphabet of a plurality of languages used in the OCR of the selected word or word combination.

6. The electronic apparatus of claim 1 , wherein the instructions further cause the electronic apparatus to:

prior to detecting the selection of a word or word combination in the image, establish a geo-location position of the electronic apparatus;

identify a region or country based on the geo-location position;

determine one or more preferred languages for the identified region or country;

generate a ranked list of preferred languages for the region or country; and

limit the character alphabet of a plurality of languages used in the OCR of the selected word or word combination.

7. The electronic apparatus of claim 1 , wherein the instructions further cause the electronic apparatus to:

display a translation of a multi-lingual word string with text of a secondary language in its original form; and

emphasize the word or words of the multi-lingual word string that are in the secondary language.

8. The electronic apparatus of claim 7 , wherein the instructions further cause the electronic apparatus to:

detect a selection on the emphasized word or words in the secondary language; and

translate the selected emphasized word or words into a target language.

9. The electronic apparatus of claim 1 , wherein the instructions further cause the electronic apparatus to:

display proposed words from the language specific word list where the proposals are based on similarity of spelling with a desired word or word combination to be translated;

detect a selection of a proposed word or correction of a variant; and

provide a translation of the proposed word or correction of the variant.

10. A computer-implemented method for translating user selected words without preliminarily specifying a source language of translation, the method comprising:

acquiring an electronic image of text;

detecting a selection of a word or word combination in the image to be translated;

performing optical character recognition (OCR) on the selected word or word combination using a character alphabet of a plurality of languages;

generating a set of recognition variants for each word of the selected word or word combination;

transmitting each set of recognition variants to a set of language specific processors;

eliminating language inappropriate variants from the set of recognition variants for each language at least by matching characters in a variant with alphabetic characters of the language specific processors;

matching each of the remaining variants to a dictionary;

translating a word variant using a translation dictionary; and

providing a translation of the word variant.

11. The computer-implemented method of claim 10 , wherein the method further comprises:

displaying proposed words from the language specific word list where the proposals are based on similarity of spelling with a desired word or word combination to be translated;

detecting selection of a proposed word or correction of a variant; and

providing a translation of the proposed word or correction of the variant.

12. The computer-implemented method of claim 10 , wherein said matching of each of the remaining variants to a specific language dictionary includes:

identifying remaining variants in the dictionary of the language based on a language specific processor.

13. The computer-implemented method of claim 10 , wherein said eliminating language inappropriate variants from the set of recognition variants includes:

eliminating an inappropriate variant based on detection of a language inconsistent character.

14. The computer-implemented method of claim 10 , wherein the method further comprises:

prior to detecting the selection of a word or word combination in the image, performing OCR on a portion of the image of text to generate recognized words;

searching for the recognized words in monolingual dictionaries;

generating hypotheses about one or more source languages of the text;

ranking hypotheses based on a frequency of each language's use in the text; and

using the ranked hypotheses to limit the character alphabet of a plurality of languages used in the OCR of the selected word or word combination.

15. The computer-implemented method of claim 10 , wherein the method further comprises:

prior to detecting the selection of a word or word combination in the image, establishing a geo-location position of the electronic apparatus;

identifying a region or country based on the geo-location position;

determining one or more preferred languages for the identified region or country;

generating a ranked list of preferred languages for the region or country; and

limiting the character alphabet of a plurality of languages used in the OCR of the selected word or word combination.

16. The computer-implemented method of claim 10 , wherein acquiring an electronic image of text includes using a camera of the electronic device or accessing the electronic image of text from a memory storage of an electronic device.

17. The computer-implemented method of claim 10 , wherein the recognized variants represent a word, word combination, abbreviation or hieroglyph, and wherein the recognized variants include characters recognized as a group.

18. The computer-implemented method of claim 10 , wherein the method further comprises:

displaying a translation of a multi-lingual word string with text of a secondary language in its original form; and

emphasizing the word or words of the multi-lingual word string that are in the secondary language.

19. The computer-implemented method of claim 10 , wherein the method further comprises:

detecting a selection on the emphasized word or words in the secondary language; and

translating the selected emphasized word or words into a target language.