The invention belongs to the technical field of video editing, and particularly relates to a method and a device for automatically editing and synthesizing a video. The method comprises the followingsteps: carrying out image and voice recognition on an initial video to obtain image frame data and audio frame data; detecting the image frame data and the audio frame data to obtain detection resultsof the image frame data and the audio frame data; obtaining a video editing instruction; according to detection results of the image frame data and the audio frame data, selecting an image frame andan audio frame which conform to the video editing instruction; and generating a video from the selected image frame and audio frame. The method comprises the steps of establishing an initial video material library, identifying an initial video, obtaining image frames and audio frames of the initial video, adding screening parameters to each image frame and audio frame, quantifying video characteristics, automatically screening the image frames and the audio frames according to the quantized video characteristics, and obtaining edited video finished products from and from the video finished products.