Spell checker for Scottish Gaelic
(The information here replaces replaces the old page
detailing Roy Wentworth’s original spell-check wordlist - which might still be of some use for old computers)
This spell-checker for Scottish Gaelic was developed under Pròiseact an Dearbhair by the European Language Initiative (TELI) with funding from Bòrd na Gàidhlig.
The Dearbhair contains over 540,000 word forms and has been developed from the original spell-checker devised by our late colleague Dr Roy Wentworth.
Note: In order to make it available to PC users ahead of a full Gaelic spell-checker being posted by Microsoft, we have developed this interim version as a supplement to Word's English spell-checker. It therefore provides users with a bilingual facility. It should be noted that to operate the Dearbhair your computer language should be set to English. Do not set the language to Gaelic (Scotland).
- Right click on the file dearbhair-utf16.dic, select ‘Save Link As …’ (or ‘Save linked content as …’ in some browsers) and save it somewhere safe on your computer (such as C:\Users\Public\)
- With a document open in Word, go to File>Options>Proofing>Custom Dictionaries>Add …
- Browse your way to the file and click OK
The Dearbhair will now function alongside the built-in spell-checker on your computer as a kind of supplement to the English.
TELI regards the first year following introduction of all its publications and products as a period of public consultation. If you encounter any forms that you believe to be incorrect or wish to propose any additions, please notify TELI by e-mail without delay at the following address: email@example.com
Some recommended settings:
"Check spelling as you type" - on;
"Check grammar errors as you type" - off;
"Check grammar with spelling" - off.
If you notice that spell-checking is not working in a document, or in part of a document,
you may have to do Edit>Select All and set the language to English.
For non-Microsoft systems, rather than UTF-16 the format you need the spellcheck wordlist in is more likely to be UTF-8: