Towards an ImageNet Moment for Speech-to-Text

Could Speech-To-Text be next?

Posted by snakers41 on March 28, 2020



So, as you may know, we have been extensively working on STT datasets and STT models.

So, without further ado - please read our article here - https://thegradient.pub/towards-an-imagenet-moment-for-speech-to-text/.

This is just a placeholder article for my RSS readers, we have done too much rewriting for thegradient and for the Russian version, that I feel tired to do it once more for the blog.

If you prefer the abridged short Russian version, then please go here https://habr.com/ru/post/494006/.

You can learn more about our dataset, Open STT here and here.