Alert icon
We're changing our privacy policy. This stuff matters.  Learn more  Dismiss

Translation by Iterative Collaboration between Monolingual U

Loading...

Sign in or sign up now!
4,575
Loading...
Alert icon
Sign in or sign up now!
Alert icon

Uploaded by on Sep 29, 2009

Google Tech Talk
September 24, 2009

ABSTRACT

Translation by Iterative Collaboration between Monolingual Users. Presented by Benjamin Bederson.

Human translation is expensive and slow, and often unavailable between uncommon language pairs. Machine translation is inexpensive and fast, but quality remains unreliable. In an effort to find a balance between speed and quality, we describe a new iterative translation process designed to leverage the massive number of online users who have limited or no bilingual skill. The iterative process is supported by combining existing machine translation methods with monolingual human speakers. We are building a publicly available prototype on the Web that is capable of yielding high quality translations at drastically reduced expense. In this talk, I will describe our translation model, and show our early prototype. This work is being done in collaboration with Philip Resnik at UMD.

Benjamin B. Bederson is an Associate Professor of Computer Science and the previous director of the Human-Computer Interaction Lab at the Institute for Advanced Computer Studies and iSchool at the University of Maryland. His research is on mobile device interfaces, information visualization, interaction strategies, digital libraries, and accessibility issues such as voting system usability. He is also cofounder and Chief Scientist of Zumobi, a mobile media company.

Category:

Science & Technology

Tags:

License:

Standard YouTube License

  • likes, 1 dislikes

Link to this comment:

Share to:

Top Comments

  • hmm...with the google books making all books every written available digitally, then this project giving the ability for massively inexpensive and scalable translation to be done on those books....then we'll need an open source kindle to deliver the context....

    then everyone in the world can read everything else written by anyone else on the planet, ever. that sure will be a lot of data ;p

see all

All Comments (7)

Sign In or Sign Up now to post a comment!
  • @fernandoscolletion: The same reason people "work" on Wikipedia for free.

  • I read a comment, why should we? Because it is the means that the world becomes one. For idealistic, humanitarian, global and no-so-altruistic reasons. BUT geez, so much of what we all do on the internet these days IS contribute, our ideas, opinions and etc, and this is how the global human family evolves. Surely enough money is made as a translator/interpreter (I know I make a nice salary, no complaints), but this opens vistas of great opportunities for the world to become a true family.

  • I love this, but it will improve depending on competent translators reviewing the software and uploading their material. BUT you know, it is relatively the beginning! Without human contributions, then the only machine translation capable would require extremely fast processors, huge RAM, and a very expensive program. Google's approach would be best, with the target language speakers to clean up translations or the source languages to do the same. Great idea.

  • If we all work in collaboration. Money would disapear with expeculation, war and economical dominance.

  • Volunteers for translating? Who is paying you for this project? Google? Why should we, translators, work for free?

Loading...

0 / 00Unsaved Playlist Return to active list
    1. Your queue is empty. Add videos to your queue using this button:
      or sign in to load a different list.
    Loading...Loading...Saving...
    • Clear all videos from this list
    • Learn more