Preferences

devnullbrain parent
>All this is not to say Google Translate is doing a bad job

Google Translate is doing a bad job.

The Chrome translate function regularly detects Traditional Chinese as Japanese. While many characters are shared, detecting the latter is trivial by comparing unicode code points - Chinese has no kana. The function used to detect this correctly, but it has regressed.

Most irritatingly of all, it doesn't even let you correct its mistakes: as is the rule for all kinds of modern software, the machine thinks it knows best.


simonw
That doesn't sound like a problem with Google Translate, it sounds like a problem with Google Chrome. I believe Chrome uses this small on-device model to detect the language before offering to translate it: https://github.com/google/cld3#readme
devnullbrain OP
Archived in 2024
simonw
Looks like it's still vendored by Chromium: https://github.com/chromium/chromium/tree/main/third_party/c...
numpad0
IMO, it's still not too late and it'll never be too late to split and reorganize Unicode by languages - at least split Chinese and Japanese. LLMs seem to be having issues acquiring both Chinese and Japanese at the same time. It'll make sense for both languages.

The syntaxes aren't just different but generally backwards, and, it's just my hunch but, they sometimes sound like they are confused about which modifies word which.

It's only a matter of time before they have an LLM both 1. cheap 2. fast 3. good enough that they'll replace Google Translate's current model with it. I'd be very surprised if they'd put more than 1 hour of maintenance into Translate's current iteration over the last 12 months.

This item has no comments currently.