This setup allows to train end-to-end neural models for spoken language understanding (SLU). - View it on GitHub