Hacker News Reader

Comment by jamestimmins | Hacker News Reader

Obviously, a lot of companies prefer to host enterprise software on-prem rather than allow them to run in the cloud.

Is anyone aware if there's been a similar movement to create on-prem versions of popular ML-based products? Otter, Grammarly, etc.

View on HN

If you are looking for a self-hosted Grammarly alternative, LanguageTool [0] might be for you.

I use LTeX [1] for VSCode, which sets up a local LanguageTool server, and its resource usage is quite significant. (I use it together with their n-gram data sets [2])

[0] https://dev.languagetool.org/

[1] https://github.com/valentjn/vscode-ltex

[2] https://dev.languagetool.org/finding-errors-using-n-gram-dat...

A somewhat unrelated question: would you happen to know any libraries for autocomplete suggestions based on n-grams from a custom corpus?

For offline transcription try Vosk https://github.com/alphacep/vosk-api

+1 for Vosk. If you want an easy way to get started with it, try mp4grep [1] which is a ready-made program you can unzip and run. To transcribe a whole file, simply:

  mp4grep --model ~/apps/mp4grep-0.1.1/model/ --transcribe filename.mp3

The output is more intended for captioning so it's lots of short phrases with timestamps and no punctuation, but it'll give you a quick taste of what Vosk can do.

[1] https://github.com/o-oconnell/mp4grep

Dragon can do transcription and works pretty well. It’s like $500.

Reply on news.ycombinator.com