TikTok

Contenus

• Reading interaction method, device, equipment, server and storage medium

  The invention discloses a reading interaction method, device and equipment, a server and a storage medium. The method comprises the following steps: receiving a user face image captured when a user reads; identifying the face image of the user, and determining a state expression of the user; and when the state expression meets a reading interaction condition, determining interaction content corresponding to the state expression and playing the interaction content. By means of the method, emotion change sharing between the user and the virtual reading partner in the reading process is achieved through state expression interaction with the reading user in the reading process, so that the single boring reading process is improved, the reading interestingness of the user is improved, and the reading enthusiasm of the user is improved.
• Sound effect adding method and apparatus, storage medium, and electronic device

  The present invention relates to a sound effect adding method and apparatus, a storage medium, and an electronic device. The method comprises: determining, on the basis of an emotion judgment model, a statement emotion label of each statement of a text to be processed; determining an emotion offset value of said text on the basis of the type of emotion labels which are largest in quantity among the multiple statement emotion labels; for each paragraph of said text, determining an emotion distribution vector of the paragraph according to the statement emotion label of at least one statement corresponding to the paragraph; determining emotion probability distribution of the paragraph on the basis of the emotion offset value and the remotion distribution vector corresponding to the paragraph; determining, according to the emotion probability distribution of the paragraph and sound effect emotion labels of multiple sound effects in a sound effect library, a target sound effect matching the paragraph; and adding the target sound effect to an audio position corresponding to the paragraph in an audio file corresponding to said text. Thus, the effect of automatically selecting and adding sound effects can be implemented, and the efficiency of adding sound effects can be improved.
• The method and apparatus of personage's emotion for identification

  The embodiment of the invention discloses a method and device for identifying character emotion. A specific embodiment of the method comprises the steps of extracting a face image set from a to-be-processed figure video, dividing the face image sert into at least one face image group based on a matching relationship between the face images, the different face image groups corresponding to different persons displayed by the person video, for each face image group in the at least one face image group, performing expression recognition on each face image in the face image group to obtain a corresponding expression recognition result, and determining the emotion information of a person corresponding to the face image group based on the expression recognition result corresponding to each face image in the face image group. According to the embodiment, the character emotion recognition based on facial expressions is realized.
• Video dubbing method and device, electronic equipment and computer readable storage medium

  The invention provides a video music matching method and device, electronic equipment and a computer readable storage medium, and relates to the technical field of video processing. The method comprises the following steps: inputting a video to be subjected to music matching into a pre-trained first neural network model to obtain specific-dimension video features of the video to be subjected to music matching; inputting the video features of the specific dimension into a predetermined second neural network model to obtain an emotion category of the video to be subjected to music matching; obtaining a plurality of songs corresponding to the emotion categories from a song library pool, and respectively extracting audio features of the songs; and performing Euclidean distance calculation according to the specific dimension video features of the video to be subjected to music matching and the audio features of each track, and taking the track of which the Euclidean distance is within a preset range as a recommended track of the video to be subjected to music matching. According to the invention, based on the emotion category expressed by the video, automatic music matching can be carried out on the video, and the use experience of a user is improved.
• Video file processing method, system, medium and electronic equipment

  The invention provides a video file processing method and system, a medium and electronic equipment. The method comprises the steps of obtaining voice comment information input by a user for a current video file, wherein the voice comment information comprises voice content, voice duration and comment mood; identifying the content of the current video file, and generating a plurality of video scenes; determining the video scene matched with the voice comment information; and when the video file is played to the video scene, outputting the voice content. According to the method, the interactive interestingness of the reviewer can be improved; and user viscosity can be further increased.
• Video processing method and apparatus, device and medium

  Provided in embodiments of the present disclosure are a video processing method and apparatus, a device, and a medium. The method includes: receiving a play trigger operation for a forwarded video, the forwarded video being forwarded from an original video by a forwarding user; obtaining the original video, and forwarding comment information provided by the forwarding user on the original video when forwarding the original video; and playing the original video on a play interface of video works of the forwarding user, and displaying the forwarding comment information. In the embodiments of the present disclosure, when the forwarding user forwards the original video, the original video can be played when receiving a play station for the forwarded video, and meanwhile, the forwarding comment information provided by the forwarding user on the original video during the forwarding is displayed while playing the original video, instead of only displaying the comment information of the creator as in the related art. Thus, in this case, it is convenient for the user to know the forwarding user's feeling about the forwarded video, thereby improving the user's ense of experience.
• Video recommendation method and device, electronic equipment and computer-readable storage medium

  The invention provides a video recommendation method and device, electronic equipment and a computer readable storage medium, and the method comprises the steps: obtaining the sensory information of a user when the user is detected to watch a video; sending the sensory information to a server, receiving related information of the to-be-recommended video returned by the server, and displaying the related information of the to-be-recommended video to the user, wherein the to-be-recommended video is determined based on sensory information. Scheme of the present disclosure, when it is detected that the user watches the video, the sensory information of the user is obtained, the watching state of the user, including the watching feeling of the currently watched video and the to-be-recommended video determined based on the sensory information of the user, can be reflected on the basis of the sensory information, and the watching feeling of the user is considered, so that the recommended video is more accurate and more accordant with the mind of the user.