Of this entire world having the different persons and different languages. But of course English is the common language to all around world, but some people learn their own native languages like Telugu, Hindi, Urdu, etc..... based on their interest. But they want the content or news in Google should be in their own languages.
For example, when the Google CEO Satya nadella came to India, the chief minister of Hyderabad asked him to give entire content in Google with Telugu in andhrapradesh. Of course, I'm also the one of the person from Andhra pradesh, India.
By this way, everyone wants to be in their own languages. So for that Google developed " machine translation algorithm " that changing cross culture communication through Google translate.
Google engineers have trained a machine learning algorithm to write picture captions using the same technique it developed for language translation.
That technique is now used to translate pictures into words, automatically caption the images. But in recent years, Google has worked out to translate text in an entirely different way.
You can also see : Do you know???? in what Google is working on now?????
Google goes on important assumption. This is that specific words have the same relationship to each other regardless of the language. For example, " king - man + woman = Queen " should hold true in all languages.
Google translating a sentence into vector and then using that vector to generate the equivalent sentence.
Now Orial vinyals and pals at Google are translating images into words. To test the efficiency of this approach they used human elevators recruited from Amazon's mechanical Turk to rate captions generated automatically in this way along with those generated by other automated approaches and by humans.
Google named this mechanism as " Neural image caption " , they used a well-known dataset of images called PASCAL, neural image capture.
Add more story or any ideas....???? Please comment below......
For example, when the Google CEO Satya nadella came to India, the chief minister of Hyderabad asked him to give entire content in Google with Telugu in andhrapradesh. Of course, I'm also the one of the person from Andhra pradesh, India.
By this way, everyone wants to be in their own languages. So for that Google developed " machine translation algorithm " that changing cross culture communication through Google translate.
Google engineers have trained a machine learning algorithm to write picture captions using the same technique it developed for language translation.
That technique is now used to translate pictures into words, automatically caption the images. But in recent years, Google has worked out to translate text in an entirely different way.
You can also see : Do you know???? in what Google is working on now?????
Google goes on important assumption. This is that specific words have the same relationship to each other regardless of the language. For example, " king - man + woman = Queen " should hold true in all languages.
Google translating a sentence into vector and then using that vector to generate the equivalent sentence.
Now Orial vinyals and pals at Google are translating images into words. To test the efficiency of this approach they used human elevators recruited from Amazon's mechanical Turk to rate captions generated automatically in this way along with those generated by other automated approaches and by humans.
Google named this mechanism as " Neural image caption " , they used a well-known dataset of images called PASCAL, neural image capture.
Add more story or any ideas....???? Please comment below......
No comments:
Post a Comment