[GDC 2014] What is the Place of Machine Translation in Today's Gaming Industry?
Upcoming SlideShare
Loading in...5
×
 

Like this? Share it with your network

Share

[GDC 2014] What is the Place of Machine Translation in Today's Gaming Industry?

on

  • 1,167 views

Game Developers Conference

Game Developers Conference

+ Info: http://www.gdconf.com/

Statistics

Views

Total Views
1,167
Views on SlideShare
1,167
Embed Views
0

Actions

Likes
0
Downloads
0
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

[GDC 2014] What is the Place of Machine Translation in Today's Gaming Industry? Presentation Transcript

  • 1. What  is  the  place  of     Machine  Transla2on     in  today's  gaming  industry?             Moderator:  Mike  Dillinger   President,  Associa/on  for  Machine  Transla/on  in  the  Americas     Organizer:  Michel  Buch-­‐Andersen   Director,  Business  Development,  Pole  To  Win  Interna/onal      
  • 2. MT  from  the  outside       Transla2on  Engine   Input  Sentence   User   Translated  Sentence   Human  Review     Proposed  Transla2on  or  No  Match   TM  MT  +     Prepara2on:   Build  TMs  with   relevant,  matching   sentences   Prepara2on:   “Train”  MT  engines   with  relevant,  matching   sentences   Usefulness:   Be?er  match  between   TM  data  and  input  =   more  useful   Usefulness:   Be?er  match  between   MT  data  and  input  =   more  useful   Final  Quality:   Machine    +    Human   Final  Quality:   Machine    +    Human   ✗ No  training!   ✗ No  review!   ✗ Machine  Only!   Machine  Transla/on  adds  extra  func/onality  for  translators.  
  • 3. MT  from  the  inside       Transla2on  Engine   Input  Sentence   TM  MT  +     Full-­‐segment  Match   •  Match,  Fuzzy  Match,  No  Match   •  Assembled  Transla8ons?   Known  sentences  Only   Human  Quality   Piece-­‐by-­‐piece  Match   •  Match,  Fuzzy  Match   •  Assembled  Transla8ons   Unknown  sentences,  too   Variable  Quality   Proposed  Transla2on   Other  sources  of   Sample  Transla/ons:   “corpora”  
  • 4. Ques/ons  for  Discussion     Transla2on  Engine   Input  Sentence   User   Translated  Sentence   Human  Review     Proposed  Transla2on  or  No  Match   TM  MT  +     Prepara2on:   “Train”  MT  engines   with  relevant,  matching   sentences   Usefulness:   Be?er  match  between   MT  data  and  input  =   more  useful   Final  Quality:   Machine    +    Human   1:    How  useful  is  MT   right  now?   2:    Which  MT  system   should  I  use?   3:    Which  content  is   best  for  MT?   4:    What’s  the  best   process  for  using   MT?  
  • 5. The  Panelists   •  Michaela  Bartelt   Senior  Director,  Worldwide  Localiza/on,  EA   •  Beatrice  Compagnon     CEO,  Vienna  Group  Interna/onal   •  Declan  Groves   Linguis/c  Engineer,  MicrosoU   •  Kir2  Vashee   VP  of  Enterprise  Transla/on  Sales,  Asia  Online