CONTEXT ACQUISITION METHOD AND DEVICE BASED ON VOICE INTERACTION

摘要

Provided in the embodiments of the present invention are a context acquisition method and device based on voice interaction. The method comprises: acquiring a scene image collected by an image collection apparatus at a voice starting point of the present conversation, and extracting a human facial feature of each user in the scene image; if it is determined, according to the human facial feature of each user and a human face database, that there is a second human facial feature matching a first human facial feature, acquiring a first user identifier corresponding to the second human facial feature from the human face database, wherein the first human facial feature is the human facial feature of one user, and the second human facial feature is the human facial feature, which is stored in the human face database, of a user in a conversation state; and if it is determined that there is a stored conversation corresponding to the first user identifier in a voice database, determining context of voice interaction according to the present conversation and the stored conversation, and after a voice end point of the present conversation is obtained, storing the present conversation in the voice database. By means of the present embodiment, precision in acquiring context of voice interaction can be improved.