Have you tried https://speechmatics.com/ ? I think they have a specially tuned medical version, and quite a generous free allowance. - Source: Hacker News / about 1 year ago

Here is one library that will be used for the training https://fasttext.cc/ this allows for the consensus across multiple languages so that we can define our mystery word correctly. Source: over 2 years ago

(response to edit) > The classification problem is interesting though. I ended up with a long list of hundreds of topics. Most articles fall in two or more. There's also a sub-problem of clustering news by subject. Yeah, certainly difficult. I'm doing it partially manually right now but also with fastText[1]. I'd like to switch completely to fastText soon though since more often than not the newsletters I add... - Source: Hacker News / about 3 years ago

I'm planning to build a business on this, so probably won't open-source it--but I'm always looking for interesting things to write about! I write a weekly newsletter called Future of Discovery[1]; I might write up some more implementation details there in a week or two. In the mean time, most of the heavy lifting is done by the Surprise python lib[2]. It's pretty easy to play around with, just give it a csv of... - Source: Hacker News / about 3 years ago

FastText is a Facebook tool that, among other things, is used to train text classification models. Unlike Tensorflow.js, it is more intended to work with text so we don't need to pass a tensor and we can use the text directly. Training a model with it is much faster and there are fewer hyperparameters. Besides, to use the model from the browser is possible through WebAssembly. So it's a good alternative to try.... - Source: dev.to / about 3 years ago