[marjipan200]

interesting, but curious how it would handle non-latin script input such as Japanese. seems like dialect.py uses regex for latin script extraction only