Content recognition and censorship

摘要

A method for controlling the output of media content, particularly audio content, to a user in relation to an apparatus, such as a set-top-box, having an input, for receiving media content, and an output, for providing the media content to an output device such as a television or other display device. Media content is received at the input and then analysed using a media content recognition unit. The media content recognition unit undertakes the step of identifying portions of the received media content that contain undesired content, such as offensive language. The output of media content to a user is then controlled, based on any identified undesired content, to prevent the output of content when the identified undesired portions are to be output. A corresponding apparatus and computer program are also provided.

主权项

1. A method for controlling the output of media content to a user, for an apparatus comprising an input for receiving the media content and an output for providing the media content to an output device, the method comprising the steps of:
receiving the media content at the input, wherein the media content comprises a plurality of words; analyzing the media content using a media content recognition unit to identify one or more undesired words within the media content based on a list of undesired words, wherein: the list of undesired words are stored by the apparatus; and a number of undesired words present on the list of undesired words is selected based on a time taken for a decoder to decode the media content; and controlling the output of the apparatus to prevent the output of the media content to the user when the one or more undesired words of the media content are to be output; wherein the media content includes audio content and wherein: the step of analyzing the media content using the media content recognition unit to identify one or more portions that contain undesired content includes comparing the received audio content with data stored on a memory, the data identifying the one or more undesired words; wherein the method further comprises performing comparisons with a maximum number of undesired words present on the list of undesired words; wherein the maximum number of undesired words is selected such that the recognition unit can compare the received audio with the maximum number of undesired words within a specified time period; wherein the maximum number of undesired words is selected based on the time taken for the decoder of the apparatus to decode a portion of the media content, the maximum number of undesired words being such that the recognition unit can compare the portion of the received audio content with the maximum number of undesired words within the time taken for the decoder to decode the portion of media content; and wherein the maximum number of undesired words is selected such that the comparison of the one or more undesired words with the received audio content can be performed at a same rate, or a greater rate, as the received audio content is decoded by the decoder; wherein the media content has an associated time reference; identifying, using the associated time reference, a beginning and an end time of each media content portion containing undesired content; and preventing the display of the media content to the user for a period of time starting at or near the beginning time identified using the associated time reference so as to prevent the output of all or some of the undesired content.