An open-source library for recognition of speech commands in the user dictionary using audiovisual data of the speaker
