We showcase here a micro-service for language identification. The algorithm is based on a naive bayes implementation multiplying n-gram probabilities, and assuming equal a priori probabilities for each languages. Inference is expected to be quite fast, basically O(K*L) with K the number of activated languages and L the length of the input text. The specification file is defined according to openapi v3 (OAS3).

You may select different options below to test it out.