Dolphin: A Large-Scale Automatic Speech Recognition Model for Eastern Languages
https://arxiv.org/abs/2503.20212Open linkView original on lemm.eeSyndicated from the fediverse. Read and engage on the original instance.
View original on lemm.eehttps://arxiv.org/abs/2503.20212Open linkView original on lemm.ee
2 replies
Guess 'Diversity' in language wasn't important for neither the Anglophone world, or Saltman. Good for Asia, but afaik we still lack descent support for Africa, the middle-east, and a shitload of smaller languages that Western corps didn't bother adding.
Technically it supports fewer languages than whisper, 40 vs 99
The main problem isn't "bother", it's training data. You need hundreds of thousands of hours of high quality transcripts to train models like these and that just doesn't exist for like zulu or whatever