I am building a web application that allows users to transcribe, word by word, Early Modern English manuscripts.

I want to tell users when a word they transcribed may be a new spelling variant of that word.

For doing that I need to interrogate a database via an API call. The database would be a comprehensive list of spelling variations of all words in Early Modern English.

I've tried the Oxford English Dictionary one but it seems a bit unreliable (sometimes it takes 40+ sec to return just a few records).

Has anyone done this before? Does anyone know of any API that could help?


  • I think you'll find more people with expertise in NLP over on our sister site, Linguistics. The EL&U regular I'm aware of who might be able to shed light on this topic is @tchrist, but I haven't seen him around in a while. – Dan Bron Jun 8 '15 at 15:12
  • 1
    Depending on the usage that you envisage it might be worth taking a look at Vard, ucrel.lancs.ac.uk/vard/userguide, as it seems to be doing much of the same thing. – Steve Barnes Jun 8 '15 at 19:04
  • @SteveBarnes thanks. I knew about Vard, but I thought it was a desktop based application. I need something I can access from the web. Or am I missing something? – U r s u s Jun 8 '15 at 19:59
  • That is what I meant about depending on usage - IIRC Vard is Java based so it should be possible, with the permission and cooperation of Alistair Baron and Lancaster University, to split out the back-end processing to run on something like an Amazon service if your usage is academic. Then you would benefit from what they have done and possibly Vard could get the logs from the service to expand the corpora. This would be win-win. – Steve Barnes Jun 9 '15 at 5:24
  • @SteveBarnes Usage will be academic. It's for a citizen science project. Wouldn't that be a lot of work for Alastair? – U r s u s Jun 9 '15 at 9:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.