Abstract:
A management of user profiles comprises calculating, for each speaker model of at least one speaker model, a confidence measure representing a probability that the speaker model represents a speaker of a cluster of audio segments. A user profile associated with the speaker model is updated based on a user preference assigned to the cluster of audio segments if the confidence measure calculated for the speaker model represents a probability that is higher than a target probability. The embodiments achieve an efficient user profile management in a voice-controlled context but without the need for any dedicated enrollment sessions to train speaker models.
Abstract:
A method and apparatus for a first device to determine profile information are described including receiving input from a second device, wherein at least one of the first device or the second device detects physical characteristics of people present in an area and the physical characteristics are used by the device to determine who is present in the area, retrieving profile information of the people present in the area, determining a relationship between profiles of the people in the area, applying rules to negotiate a compromise regarding device settings responsive to the relationship between profiles of the people present in the area and adjusting settings of a first device or a third device responsive to the compromise.
Abstract:
A method and apparatus for adjusting device settings on a first device are described including receiving input from a second device and adjusting the device settings or settings of a third device responsive to the input.
Abstract:
According to aspects of the invention there are provided a method for distributing content from one or more content sources over a communications network to a plurality of users by a content distribution system, the content distribution system comprising a master node and a plurality of local nodes in communication with the master node, each local node serving a group of users from the plurality of users, the method comprising: maintaining, at the master node, a content directory associated with a plurality of content items from the one or more content sources, the content directory comprising content location data for the plurality of content items and information associated with each of the content items; generating, at each local node, user content location data for each user in the group of users based on selecting a plurality of content items from the content directory according to a user profile of each user, the user content location data for each user including locations of the selected content items; and transmitting, by the local node, the user content location data for each user to a content player associated with said each user for use in retrieving the corresponding content items from the one or more content sources for seamless playback of the selected content items.
Abstract:
According to aspects of the invention there are provided a method for modifying a content stream distributed to one or more users, the method comprising: generating a content stream comprising a plurality of content items from one or more content providers; constructing one or more filters for modifying the content steam, wherein the one or more filters is constructed by defining one or more parameters relating the content items; applying the one or more filters to the content stream; selecting one or more content items from the content stream based on defined parameters; and modifying the content stream by applying the constructed filter to the plurality of content items in the content stream.
Abstract:
A method and apparatus for providing targeted advertisements is provided herein. In particular, targeted advertisements are provided to users based on face clustering for time-varying video. During operation video is continuously obtained of users of the system. Users' faces are detected and measured. Measurements of users' faces are then clustered. Once the clusters are available, advertisements are targeted at clusters rather than individual users.
Abstract:
본 발명은 콘텐츠에 포함된 장면에 대한 사용자의 반응에 관한 정보를 분석함으로써, 장면에 대한 사용자의 감정 상태에 관한 정보를 제공하거나 콘텐츠의 각 장면별로 부가되어 있는 정보를 선별적으로 제공하여 인터렉티브한 서비스를 제공할 수 있게 해주는 영상표시기기, 영상표시시스템 및 그 사용자 감정 상태 분석 방법에 관한 것이다. 이를 위해 본 발명의 실시예에 따른 사용자 감정 정보 분석 방법은, 콘텐츠에서 식별 정보가 부여되어 있는 장면을 출력하는 단계; 상기 장면에 대한 사용자의 반응과 관련된 정보를 수신하는 단계; 상기 사용자의 반응과 관련된 정보에 근거하여 상기 장면에 대한 상기 사용자의 감정 상태를 판단하는 단계; 및 상기 판단된 감정 상태에 상기 식별 정보를 연결하여 저장하는 단계를 포함하는 것을 특징으로 한다.
Abstract:
A system comprising at least one processor, at least one sensor electronically connected to the at least one processor, at least one means for electronically connecting at least one broadcasting device to the system, wherein the at least one means for electronically connecting at least one broadcasting device is electronically connected to the at least one processor, and computer executable instructions readable by the at least one processor and operative to use the at least one sensor to determine how many persons or the identity of a person that is capable of perceiving a content broadcasted through at least one broadcasting device electronically connected to the system and to control the content based on how many persons or the identity of a person that is capable of perceiving the content. A corresponding method and a computer readable medium having computer executable instructions for performing the method.
Abstract:
In bidirectional broadcast, information associated with broadcast is received by simple operation. As marketing information, user information other than a head of family viewing is also acquired. In addition to the viewing ID of each reception device, personal attribute information for each user ID is registered in a database. A program enabling communication with the reception device is downloaded into the information terminal containing the attribute information and the information can be input to the reception device with a simple input operation. The information input is transmitted to a server after correlating the viewing ID and the user ID. In the server, the database is searched and it is possible to know what kind of attribute the user has who has performed input and what kind of inputs the user has performed. Thus, it is possible to distribute information linked with the broadcast separately to each information terminal. Moreover, it is possible to acquire extremely effective marketing information.
Abstract:
A system for calculating an audience rating comprises a settop box (50) and a monitor (60) for providing users with programs broadcast from a broadcasting station (40), and mobile stations (10a-10c) which are accommodated in a mobile data communication network (20) and in which mobile station IDs (identification information) for specifying users. The mobile stations (10a-10c) transmit a control signal containing the mobile station ID of the settop box (50) to the settop box (50). The settop box (50) holds the mobile station IDs and viewing record and transmits them to an audience rate calculation server (70). The audience rate calculation server (70) generates audience rate data by attribute of the users from the received mobile station IDs and viewing record.