Nandha on Nostr: Machine translation is becoming really, really good of late. Conversational Tamil is ...
Machine translation is becoming really, really good of late. Conversational Tamil is getting picked up easily and subtitled in English in near real time. I would have expected that from European languages that have huge datasets, but doing it for a low resource language like Tamil, with a very few openly available datasets is quite incredible. Although it was Google doing it, it bodes well for the larger open source movement.
Published at
2024-08-22 15:01:48Event JSON
{
"id": "960160bba241ce9cbae98b531eb838e00e5bbec4bcd42c4df5acf6e0e8555922",
"pubkey": "5a0b8b882152bc9c870724079d64688c1d1157c88be45e7ba12d335077fc6d6b",
"created_at": 1724338908,
"kind": 1,
"tags": [
[
"proxy",
"https://mastodon.social/@ionhandshaker/113006274721459950",
"web"
],
[
"proxy",
"https://mastodon.social/users/ionhandshaker/statuses/113006274721459950",
"activitypub"
],
[
"L",
"pink.momostr"
],
[
"l",
"pink.momostr.activitypub:https://mastodon.social/users/ionhandshaker/statuses/113006274721459950",
"pink.momostr"
],
[
"-"
]
],
"content": "Machine translation is becoming really, really good of late. Conversational Tamil is getting picked up easily and subtitled in English in near real time. I would have expected that from European languages that have huge datasets, but doing it for a low resource language like Tamil, with a very few openly available datasets is quite incredible. Although it was Google doing it, it bodes well for the larger open source movement.",
"sig": "e24c347ebb2567b4fc38071b5758ae6e2b917fa115d14996939439166dcf96e84cb2afdaa2567e9a9876df8577cc64a6bb2aabfe9ff12f63c2ece7fd8b78ccb7"
}