Software Application for Ranking Language Translations and Methods of Use Thereof

• US 20120265518A1
• Filed: 04/16/2012
• Published: 10/18/2012
• Est. Priority Date: 04/15/2011
• Status: Active Grant

• Current Assignee: Andrew Nelthropp Lauder
• Original Assignee: Andrew Nelthropp Lauder
• Inventors: LAUDER, Andrew Nelthropp
• Granted Patent: US 8,849,628 B2
• Time in Patent Office: Days
• US Class Current: 704/3
• CPC Class Codes: G06F 40/51 Translation evaluation

Abstract

The software application is used to rank language translations and a method to implement those ranked language translations. The language translations are for textual or vocal phrases that a user wants to convert from its original language to a preferred language. The software application will refer to a variety of translation sources in order to create a set of applicable translations for a textual or vocal phrase. The software application will then rank each applicable translation with an accuracy score, which is weighed by three factors: the commonalities between the set of translation sources, a user input process, and a linguistic mapping process. Some methods to implement the ranked language translations include using accelerometer data from a electronic communication device to operate the software application, integrating a moderator into a conversation process, using subtitles in speech bubbles for a video chat, and using location data to communicate with different contacts.

Claims

1. A method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method comprises the steps of:

providing a plurality of translation sources, wherein said plurality of translation sources includes a machine language-translation engine, a human language-translation database, a language-translation storage database, and a parallel language content database;

retrieving a message in an original language through a user interface;

compiling a plurality of applicable translations for said message in a preferred language by referring to the plurality of translations sources;

determining an accuracy score for each of said plurality of applicable translations, wherein said accuracy score for a particular translation from said plurality of applicable translations is weighed by commonalities between said plurality of translation sources, a user input process, and a linguistic mapping process;

displaying said accuracy score of each of said plurality of applicable translations with said plurality of applicable translations on said user interface;

prompting to choose a best translation from said plurality of applicable translation;

recording each of said plurality of applicable translations in said language-translation storage database, wherein each of said plurality of applicable translations comprises a total rating and a total number of reverse requests; and

continuously updating said total rating and said total number of reverse requests for each of the plurality of applicable translations.

2. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 1 further comprises the step of:

increasing accuracy score of a particular translation, if said particular translation is provided by more than one translation source.

3. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 1 further comprises the steps of:

prompting to choose between a positive rating and a negative rating for each of said plurality of applicable translations;

retrieving said positive rating or said negative rating for each of said plurality of applicable translations through said user interface;

increasing said total rating of said particular translation, if said particular translation retrieved said positive rating;

decreasing said total rating of said particular translation, if said particular translation retrieved said negative rating;

prompting to enter a free form correction, if each of said plurality of applicable translations received a negative rating;

retrieving said free form correction and storing said free form correction into human language-translation database; and

assigning said free form correction as said best translation.

4. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 3 further comprises the steps of:

increasing said accuracy score of said particular translation, if said particular translation has a higher total rating among said plurality of applicable translations; and

decreasing said accuracy score of said particular translation, if said particular translation has a lower total rating among said plurality of applicable translations.

5. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 1 further comprises the step of:

adding to said total number of reverse requests for said particular translation, if said particular translation is requested as an input phrase in the preferred language.

6. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 5 further comprises the steps of:

increasing said accuracy score of said particular translation, if said particular translation has a higher total number of reverse requests among said plurality of applicable translations; and

decreasing said accuracy score of said particular translation, if said particular translation has a lower total number of reverse requests among said plurality of applicable translations.

7. A method of implementing the ranked language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method comprises the steps of:

providing a speech recognition engine and a speech synthesis engine;

providing a plurality of applicable translations between a first language and a second language, wherein each of the plurality of applicable translations has an accuracy score;

providing a best translation from said plurality of applicable translations, wherein said best translation has highest accuracy score;

retrieving accelerometer data in order to execute a bilingual conversation process;

retrieving moderator input data in order to execute a bilingual communication process; and

displaying subtitle data on an user interface in order to execute a bilingual video chat process.

8. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 7 further comprises the steps of:

providing an electronic communication device, wherein said electronic communication device comprise an accelerometer, a microphone, and a speaker;

receiving an upright signal from said accelerometer;

receiving an upside-down signal from said accelerometer in order to activate said microphone;

retrieving a voice message in said first language from said microphone;

transcribing said voice message into a textual message with said speech recognition engine;

displaying said textual message in said first language on said user interface;

receiving a tilt signal from said accelerometer in order to convert said textual message from said first language to said second language with said best translation with said highest accuracy score;

displaying said textual message in said second language on said user interface;

converting said textual message into said voice message with said speech synthesis engine; and

sending said voice message in said second language to said speaker.

9. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 7 further comprises the steps of:

receiving a speech phrase in said first language from a first user interface;

transcribing said speech phrase into a textual phrase with said speech recognition engine;

converting said textual phrase from said first language into said second language with said best translation with said highest accuracy score;

correcting said textual phrase in second language with said moderator input, wherein said moderator input is retrieved from a moderator interface;

converting said textual phrase into said voice phrase with said speech synthesis engine; and

sending said voice phrase in said second language to a second user interface.

10. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 9 further comprises the steps of:

displaying said textual phrase in said first language, said best translation, and said highest accuracy score through said moderator interface;

prompting to accept or decline said best translation for said textual phrase in said first language; and

retrieving said moderator input through said moderator interface, if said best translation is declined through said moderator interface.

11. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 9 further comprises the step of:

directly translating said voice phrase from said first language to said second language without said speech recognition engine, said speech synthesis engine, and said best translation by retrieving said moderator input.

12. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 11 further comprises the step of:

sending said voice phrase in said first language to said moderator interface;

retrieving said moderator input through said moderator interface in order to properly convert said voice phrase from said first language to said second language; and

sending said voice phrase in said second language to said second user interface.

13. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 9 further comprises the step of:

directly translating said voice phrase from said first language to said second language without said best translation by retrieving said moderator input;

14. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 13 further comprises the steps of:

sending said textual phrase in said first language from said speech recognition engine to said moderator interface;

retrieving said moderator input through said moderator interface in order to convert said textual phrase from said first language to said second language; and

sending said textual phrase in said second language to said speech synthesis engine.

15. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 9 further comprises the steps of:

displaying a video feed between said first user interface and said second user interface;

displaying a speech bubble on said first user interface and said second user interface, wherein said speech bubble contains an activity indicator;

displaying said textual phrase in said first language on said first user interface;

displaying said textual phrase in said second language on said second user interface;

displaying said activity indicator on said first user interface, wherein said activity indicator notifies said first user interface when said second user interface is retrieving said voice phrase or when said textual phrase is being translated; and

displaying said activity indicator on said second user interface, wherein said activity indicator notifies said second user interface when said second user interface is retrieving said voice phrase or when said textual phrase is being translated.

16. The method of ranking language translations by executing computer-executable instructions stored on a non-transitory computer-readable medium, the method as claimed in claim 9 further comprises the steps of:

searching for a plurality of close proximity contacts, wherein each of said plurality of close proximity contacts includes a location data;

prompting to choose a contact from said plurality of close proximity contacts; and

connecting said first user interface to said contact with said second user interface.