KantanLibrary is a repository of stock data, available for free to all KantanMT users and consisting of hundreds of datasets categorised according to language pair and domain.
Currently, KantanLibrary is available in the following domains:
- Automotive
- Financial
- General
- Legal
- Medical
- Patents
- Technical
- UN (United Nations)
Additionally, within KantanLibrary, KantanMT offers a set of general dictionaries for the most common language pairs. These can be found in the category Dictionaries.
You can add KantanLibrary datasets directly on top of your engine's training data.