The Language Technology Resource Portal is a collection of Welsh language technology resources developed here at the Language Technologies Unit. We have created these resources to be free and open ‘building blocks’, for use by Welsh coding clubs, hackers, enthusiastic volunteers, software companies (local and international) as well as researchers to create digital products in Welsh.
An important element of the The Language Technology Resource Portal is that it will provide support users with tutorials, as well as offering ideas for various products including apps, games, projects for your Raspberry Pi or enhancements for websites.
Here are the resources which may now be accessed from this website:
Corpora
- Corpus of Welsh Language Tweets
- Corpus of Welsh Language Facebook Texts
Translation
- Welsh<>English Machine Translation with Moses SMT
- Welsh<>English Aligner
Speech
- Welsh Text to Speech Voice (through the API service below)
- Paldaruo sound files
- GALLU acoustic models
API Services
- Cysill Ar-lein (Welsh spelling and grammar checker) API
- Welsh Parts of Speech Tagger API
- Welsh Text to Speech Voice API
- Language Detection API
Widgets
- Cysill Ar-lein (see also above)
- Vocab
Further details may be found in the relevant sections, together with links to the resources, and some ideas for use. Some of these resources are also available from the Github repository, and a list of these may also be found in the appropriate section.
In order to show the resources to the public, we hosted the conference Trwy Ddulliau Technolegol (Through Technological Means) on the sixth of March 2015 at Bangor University. Following this, on the seventh of March, we hosted the annual Hacio’r Iaith (Hacking the Language) event in the Management Centre also in Bangor, which provided us with an opportunity to come together in a relaxed, informal manner to share information and experiances, and to learn from each other.