690
The reason real-time translation is difficult: sometimes you have to wait for the end of the sentence.
(lemmy.world)
this post was submitted on 21 Mar 2025
690 points (99.6% liked)
interestingasfuck
6975 readers
882 users here now
interestingasfuck
founded 2 years ago
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
This Japanese interpreter did a TEDx talk about her work. She mentions a few issues with going between Japanese and English, like how subjects in Japanese are often dropped from sentences, so she once made the assumption to give a CEO a male pronoun only to find out that the CEO was female when she walked in the room shortly after.
The interpreter also says that you can't wait to have all the information about a sentence to start translating, so she likens it to "watching a thriller" because you don't know whether the verb at the end is "going to negate the whole sentence".
https://youtu.be/P-ggxpMY9q0?t=143
What if you take the speculative execution strategy and have multiple interpreters translating every possible semantic branch and then throwing out the recordings of the interpretations that were incorrect? ๐
Why would you do that?
You have to wait for the last work in the sentence anyway so doing extra calculations to throw out later isn't making it faster.