Automated assistant to deal with multiple age groups and / or vocabulary levels

摘要

A technique for enabling automated assistants according to the detected age layer of the user involved with the automated assistant and / or "vocabulary level" to adjust its behavior is described herein.In various implementations, data indicating the user's utterance may be used to estimate one or more of the user's age and / or vocabulary levels.The estimated age / vocabulary levels may be used to affect the various points of the data processing pipeline used by the automated assistant.In a variety of implementations, the point of the data processing pipeline that may be affected by the user's age / vocabulary level is an automated assistant call, speech to text (STT) processing, matching of intention, resolution of intention (or execution), natural language generation And / or one or more of text-to-speech ("TTS") processing.In some implementations, the threshold of one or more tolerances associated with one or more of these points may be adjusted, such as grammatical tolerance and lexical tolerance.