Personalized segmentation of media content may be performed to enable users to create end points to segment a media item based on user preferences, such as create a segment of a document that takes a predetermined amount of time to consume or read. In some embodiments, techniques may create new chapters or segments within existing chapters. The techniques may also create titles for the segments or chapters, possibly by leveraging text within the segment or chapter. The end points, segments/chapters, titles, and/or other information may be stored as annotations, which can be selectively used when rendering the media item to reveal the segments/chapters and other data.