The present invention discloses a device for sign language intelligent interaction capable of real-time identification of gestures. The device comprises a network television, a processor, a display and a camera. The processor comprises an obtaining unit, a processing unit, an edition unit and a decoding unit. The obtaining unit is configured to obtain audio information of the network television and viewers' sign language gestures' image information and depth information; the processing unit is configured to convert the audio information to text information; the edition unit is configured to distinguish different sound sources of the audio information and employ a virtual character sign language edition technology to edit corresponding virtual character sign language animations according tothe sound sources; and the decoding unit is configured to perform decoding processing of the image information and the depth information and convert the image information and the depth information after decoding to the text information. The device and the method for sign language intelligent interaction capable of real-time identification of gestures can help the deaf-mute with understanding of television programs and also can help the deaf-mute with communication with the television programs.