WO2018150412A1

WO2018150412A1 - Media content server, user device and method for streaming structured media content

Info

Publication number: WO2018150412A1
Application number: PCT/IL2017/050204
Authority: WO
Inventors: Ronen SHOVAL; David Braginsky; Yair Shapira
Original assignee: Blink Ap Ltd.
Priority date: 2017-02-16
Filing date: 2017-02-16
Publication date: 2018-08-23

Abstract

A server system for streaming media content includes at least one processor, storage coupled to the at least one processor, said storage storing a media file and a metadata file, the media file being playable on a media player to present a plurality of media content items, the metadata file comprising timing data that allows determination of respective start and end times of the media content items according to a predetermined sequence, and a respective communication module coupled to each processor for communicating with a user device and being configured to convey the metadata file to the user device, to receive from the user device timing data of requested media content items extracted from the metadata file and to stream the requested media content items to the user device.

Description

MEDIA CONTENT SERVER, USER DEVICE AND METHOD FOR STREAMING STRUCTURED MEDIA CONTENT

FIELD OF THE INVENTION

This invention relates to streaming of media content by a server to one or more user devices.

BACKGROUND OF THE INVENTION

The Internet is fast becoming the main medium of choice for downloading news and other video content. News providers such as online news broadcasting companies such as CNN, BBC and so on stream news over the Internet as do newspapers and other providers. Increasing numbers of people surf these websites and download news to their personal computers or smartphones. The present invention is primarily concerned with video on demand (VOD) streaming of video content and will be described by way of example with regard to streaming of news items.

Video streaming allows a host to send to a requesting user device an instance of a video in small manageable chunks without the need for the web browser to download the complete video. As the video is streamed, small chunks of the video are buffered and played after which they are replaced by new content so that at any given time only a small fraction of the complete video need be present at the user device. Video players display a timeline correlated to the duration of the complete movie and as the movie is played, a pointer moves along the timeline to show the temporal position in the complete video movie of the currently displayed segment. The timeline also serves to allow the user interactively to stream sections of the movie that are not currently buffered. For example, if during one viewing session a user only manages to see the first 30 minutes of a long movie, he can connect to the host website in a subsequent session and move the pointer along the timeline to the 30 minute point. The user device conveys this to the host, which then knows that the user device wants to receive video starting at the selected point on the timeline and streams video from this point onward to the user device.

While this is true in theory it works only coarsely in practice for two reasons. First, the timeline does not provide an exactly calibrated scale of where the currently buffered portion of the movie is located relative to the complete movie. At best, the timeline allows estimation where the user knows the complete duration of the video and so can guess more-or-less where a specific item of interest is located along the scale. Secondly, video streaming is also used extensively to convey media where the relative temporal location of specific items is, in any case, unknown a priori.

This problem is addressed in DVDs, which may contain multiple episodes each divided into multiple scenes. A menu allows the user to select which episode to display and, depending on user selection, may play the complete movie or only a selected scene. In effect, the starts of the episodes and of the constituent scenes are bookmarked by the manufacturer, allowing the user to skip to a selected bookmark.

Bookmarking is also known in the context of video streaming and addresses some of these problems. US 20090205006 discloses the use of digital bookmarks for digital content allowing a user to perform seek, fast forward, fast reverse, jumping to a scene in a movie, and so on. This finds application in "Movies on demand" where users are otherwise forced to do a "blind seek" of the content in which they are interested because the digital content is not encoded with a navigation scheme. The approach disclosed in this publication requires that the end-user insert bookmarks and so can only work when the user knows to which segment in the main movie each bookmarks refers. This, in turn, requires that the user has seen and is familiar with the video content. It is therefore unsuited to video streaming of previously unviewed content.

US 20140282681 discloses a system for chapterized streaming of video content.

A user device may pre-cache remaining chapters to allow the user an ability to skip past undesired chapters to reach the chapters they wish to currently view without wasting their time, congesting network backbone infrastructure, or placing load on the host servers. In instances in which the user wants to re-watch a scene, the user may skip backward to see the chapter again rather than consuming bandwidth and increasing server load for this activity. A chapter filter module may decide whether or not to cache, display, or provide content based on rating, tags or other associated data. Alternatively, a user may configure the device to show all chapters with a given set of criteria. Content is presented in the form of short clips or "trailers" of individual movies for selection or review of those individual movies. The trailers relate to fixed content that is stored discretely by the host, each movie being a separate file, which is only conveyed to the user upon demand.

US 20140156681 discloses a personalized content delivery computer system, principally within the field of streaming data such as articles and news briefs in a social network. Of particular relevance is the description of Fig. 10 representing a user interface for managing stream content. A stream of matched content may be presented initially in a collapsed state, but can be expanded to show other items by clicking on an expand button. This appears to be analogous to using thumbnail images to point to file folders in a computer operating system. If the folder contains images, then the first image could be used as a representative image.

In the television industry it is common for a publisher to produce programs that are ready to broadcast. In order to apply prior art techniques to allow for only partial transmission or streaming based on end-user selection, broadcast networks bookmark content and respond to user interaction to determine what to send. It is known to use playlists to accomplish this whereby different video segments are streamed, allowing different content to be watched by different users. It is also known to change the playlist in real time according to user interaction.

It is emphasized that all of the prior art so far as is known and understood relates to streaming media content that is either stored as discrete video segments by the host or that is bookmarked by the user. In the first case, the content is typically not linear. Linear content refers to content whose items are arranged sequentially in a specific order and are streamed as a continuous whole in an ordered sequence that is predetermined by the host so as to maintain this order. This is distinct from approaches where the user determines the order in which items are streamed, since in this case any initial order imposed on the structure of the media content is lost. This distinction will become clearer by reference to Figs. 3a and 3b. Fig. 3a is a graph showing video content that is streamed according to a hierarchy that has a root node N₀ connected by edges to nodes Nj and N₂. Each node is likewise connected to another node by one or more edges that represent alternative content that is selected by the user according to a menu that is displayed in conjunction with the user's media player.

This scheme is representative of the type of structure employed in US 20160323608, which discloses systems and methods for nonlinear media playback using a linear media player. A video tree representing a branching video presentation is provided, and a dynamic playlist is provided to a linear video player. A controller component interacts with the linear video player through a standard interface of the linear video player. During playback of a video content segment in the dynamic playlist by the linear video player, an indication of a user interaction with the playing video content segment is received by the controller component and from the linear video player via the standard interface, and the dynamic playlist is modified based on video content segments in a branch of the video tree. The video tree can be formed by nodes that are connected in a branching, hierarchical, or other linked form. Upon reaching a point during playback of the video where multiple video segments branch off from a segment, the user can interactively select the branch or path to traverse and, thus, the next video segment to watch.

By such means the user can watch a video that begins from a starting node in the tree and proceeds along connected nodes in a branch or path. Upon reaching a point during playback of the video where multiple video segments branch off from a segment, the user can interactively select the branch or path to traverse and, thus, the next video segment to watch.

In contrast to non-linear content, linear content implies that during both editing and playback the content is sequential i.e. the order of video frames is not changed and the video is played from some start time to some end time sequentially. This can be understood with reference to Fig. 3b, showing a graph where each node represented by an arrow is connected to the next node in sequence by a single edge. In advancing from one node to another the content is therefore linear and follows the structure established by the host. In addition, at each node there is shown a circle in dashed outline that may be traversed in whole or in part by the user. This represents a specific type of linearity that is characteristic of the invention and will therefore be discussed in greater detail below.

The video items may be stored in a single file that is accessed via a single URL but content items may be distributed across the web, in which case a controller will use metadata to determine where each video segment is located and will access the segments and stream them in proper temporal disposition. This is conceptually not unlike how a computer operating system uses a file allocation table (FAT) to access different sectors of a file and presents them to the user as a continuous whole.

US 20110200116 discloses systems and methods for seamless assembly of video/audio segments wherein during streaming/online progressive download of media, a second segment is downloaded to a client during the presentation of a first segment. The first segment is then attached to the beginning of the second segment where no jitter or gap results with the transition point either in the video or audio portion of the segments.

Fig. 1 depicts a typical prior art streaming arrangement and shows at the top of the figure part of a video file having four distinct episodes starting at times to, tj, ¾ and ¾, respectively, underneath which are four different timelines showing user interaction. The mid-section of each episode is denoted by a blank film, which denotes that this is only a small representative part of the complete video. The timelines correspond to the seek bar of a media player, whereby the user can move a pointer along the timeline to stream different parts of the video. The user is denoted by emoticons, which indicate the user's level of contentment. Thus, initially at time ¾ the user happily streams the whole of Episode 1 , which is followed at time tj by Episode 2. He watches this for a short period until time ¾ when he concludes that it is of no interest. So he moves the pointer blindly along the timeline to time t₅, in the hope of seeing the next episode. But to his chagrin he finds that he is still watching Episode 2. Alternative seeks are shown below starting at time t In one scenario he jumps blindly to time %, somewhere in the middle of Episode 3 and, having missed the beginning, is totally disoriented. Likewise, in the next scenario, he jumps from time ¾ to time t₇, with a similar result: he's somewhere in the middle of Episode 4 and has no clue as to what is happening. Moreover, he's not even aware of skipping the whole Episode 3. It would clearly be of benefit if, like the newspaper reader, the Internet user also could decide in real time "on the fly" whether to delve deeper into a news headline and display the expanded content or, alternatively, to skip to another headline, but to do so in a structured manner than prevents missed content. He would then be able to obtain news over the Internet in a more personalized format.

SUMMARY OF THE INVENTION

It is therefore an object of the invention to provide an improved content viewing experience, which allows a user to receive lead items relating to different media content and to make spontaneous decisions in real time as to whether to skim over an item, skip forward or backward, or view related content in depth.

This object is realized in accordance with an embodiment of the invention by a method and system having the features of the respective independent claims.

The invention relates to video on demand and does not require that any changes be made a priori to the video content itself. All that is required by the broadcaster is to prepare a table of metadata relating to the start and end times of selectable content, which can either be streamed or omitted as required. Each user is first streamed a copy of the metadata file, which is stored and used by the device' s media player to stream the appropriate content from the host. If the user elects to view the whole content, full continuity of the broadcast data is preserved. Alternatively, the user can skip items, in which case the video is streamed gaplessly to a different part of the video.

This will be explained by reference to Fig. 2, which depicts the same video file as described above with reference to Fig. 1 having four distinct episodes starting at times to, tj, ¾ and ¾, respectively, underneath which three different timelines show user interaction according to the invention. By default only a lead item constituting a brief introduction of each episode is streamed, followed by expanded content only if the user specifically requests. Thus, initially only a lead item hj of Episode 1 is streamed during the display of which the user does not interact with the user interface. As a result, this is followed by a lead item of Episode 2, from which the user immediately determines that this is of no interest. Therefore, at time ¾ during the display of ½ the user right- swipes the user interface with the result that streaming of ½ ends at time ¾ and is followed by ¾, which starts at time ¾ and provides a lead item of Episode 3. It is seen that the time line of the video is blank between ¾ and ¾, since the video content between these times is omitted. This time the user is interested and so at time t₅ during the display of h₃ the user down-swipes the user interface, whereupon at time t₃ when the lead item /¾ has been completely streamed, the expanded content is streamed and displayed continuing on from ¾. In an alternative scenario shown by the bottom timeline, the user right-swipes at time ¾ so as to end ¾ at time t₂ and commence display of ¾ whereupon the user right-swipes again so as to skip directly to ¾ showing the lead item of Episode 4. These scenarios are illustrative only. Swiping right jumps immediately to the next lead item as obtained from the metadata file. If the user swipes left, then the currently displayed item is aborted and is immediately followed by the previous lead item. In all cases, the current item, which may be either a lead item or an expanded item, is aborted mid-stream, and is followed by the next or the previous lead item depending on the swipe direction. Multiple swipes will skip multiple items, either forward or backward but always along a linear graph as shown in Fig. 3b. In the case where the requests that the expanded content is streamed, the current lead item completes and is then immediately followed by the expanded content. This is represented in Fig. 3b by the circles in dashed outline where at the end of a lead item, the associated expanded content is streamed either until its completion or until it is aborted mid-stream when it is either followed by the next lead item whether by default or if the user swipes right, or alternatively is followed by the preceding lead item if the user swipes left. It will be understood that Fig. 3b is schematic and is intend to show the linearity of the lead items. It is possible to jump over lead items by swiping forwards or backwards more than once in quick succession. Likewise, it is possible to abort an expanded item in the middle and revert to the next lead item in the sequence. But even in those cases where lead items are skipped, the user in effect traverses along the linear time line shown in Fig. 3b, jumping from one lead item to another in order until he reaches the one that is to be streamed. In this sense, the selection and presentation are linear rather than hierarchical as shown in Fig. 3a.

There is no guesswork on the part of the user because video content is streamed as dictated by the metadata file, which is streamed to the user device and stored in memory prior to streaming any of the actual video content. This obviates the need for blind seeking on the part of the user and also allows use of a media player that has no seek bar, although one may be provided if desired. In some embodiments, the seek bar is omitted altogether for the lead items but is nevertheless provided for the expanded content. Moreover, since the metadata is compiled at the server-end and is conveyed to the user device before streaming the media content, the user is able to skip from one item to another in either direction without the need to view the content and add bookmarks at the client-end as is known in the art.

As a result, the invention is also applicable to dynamic content that is subject to constant changes and updates, for example, news bulletins. These are frequently presented in the form of lead items that relate the news in brief followed by expanded broadcasts that relate the news in detail. In a typical arrangement, a news broadcast contains stories. Each story forms a continuous whole that follows a previous story and precedes a subsequent story without there being any discernible gap between consecutive stories. We use the term "video item" to denote a story but the term "story" depicts more clearly that the video content behind this item is a self-contained whole and that the broadcast containing multiple stories back-to-back is likewise a self- contained whole. A user may lack the interest or the time to watch the complete broadcast. When he streams the broadcast over the Internet either to a smartphone or to a computer, it is easy to skip content but unless the video is bookmarked by the host and the bookmarks are displayed to the user, the only option is to do a "blind seek" as discussed earlier. News stories commonly commence with a lead item that identifies the nature of the story and allows the viewer to determine whether or not it is of interest. After the lead item, the story is related in expanded detail but the expanded content and the lead item form a continuous whole. This is altogether a different structure to many of the prior art scenarios discussed above, where different media items are discrete, self- contained structures. To put this into better perspective, in the present invention the lead item together with the expanded content form part of a unitary item. During playing of the item, there is no discernible gap or hiatus between the two that defines this as the lead item and that as the expanded content. Rather somewhere along the timeline of the lead item it becomes possible for the user to make the assessment to watch the whole story or to skip to the next story.

As noted above the server is passive and takes no part in the decision as to which media content to stream. Media items are streamed from the server according to seek times that are determined by the user device from the metadata and conveyed to the server with the seek message. The cycle then continues. To this extent, the lead items are conceptually similar to newspaper headlines, each of which is designed to present the gist of an article allowing the reader to determine substantially instantaneously whether the article is of interest and, if so, to read the article in full, or to skip to a different headline if not. But there are two highly significant differences. First, a newspaper headline is quite separate from the body of the story: it is not itself part of the story but is placed afterward to attract attention. Secondly, whereas a reader has the freedom of choice to skip headlines when reading a newspaper, a person downloading news via the Internet does not. This means, that the Internet reader has no choice but to display and read or listen to the entire news content, regardless of whether or not it is of interest; or to blind skip to a different story. In contrast, in the invention as explained above, the headlines and the expanded items form a unitary item.

It should also be noted that in applications of the kind addressed by the present invention, multiple user devices are coupled to a passive server that simply holds and streams content. This means that the server does not actively dictate what content is streamed but responds to timing information received from the user devices for streaming the associated media content. Typically, the server is realized by multiple servers that are operated by cloud computing services in known manner and which are distributed. Data pertaining to a media file whose content is to be streamed to a linear media player need not even be stored in the same server.

In accordance with one embodiment of the invention, a user obtains an on-line news experience that more closely resembles the reading of a newspaper where the reader likewise can determine in real time whether a headline warrants further investigation or whether to skip to another article. It should be noted that in the context of the description and claims, the term "real time" relates solely to the user's perspective. The media content is pre-processed by the media host. The pre-processing does not need to make any changes to the media content: indeed this may even be forbidden by the content provider or publisher. Rather, the pre-processing encompasses creating metadata that minimally defines the start and end times of the different stories. Optionally, such pre-processing may take into account user preferences and ensure that media content already viewed by a user is not streamed again in a subsequent session. The start and/or end times of different parts of the video may be determined automatically, but for the purpose of the present disclosure it will be assumed that the times are entered manually into the metadata file.

The ability to determine whether to view lead items or stream a complete story virtually instantaneously in real time is considered to be of great importance. User- profiling is commonly used in a multitude of Internet platforms to allow an IP host to send different content to different end-users depending on their profiles. For example, the IP host will send different content to a 30-year old interested in politics and sports than to a 50-year old interested in science and technology. News hosts can likewise stream different media content based on user profiles as disclosed, for example, in US 20150020106. But this in effect means that the host predetermines the user's experience and removes the essential element of control and indeed of spontaneity that is characteristic of news and lends the term its name. For 'news' implies novelty: something that is not yet known, while profiling relies on prior knowledge to determine future behavior. Therefore to present news in a manner that does not allow a user who may, for example, not previously have exhibited any interest in boxing and to whom news items relating to boxing are pre-filtered by the host, nevertheless to be informed that Muhammad Ali passed away and to be given the choice to receive expanded content relating to his career and philanthropy is to deprive the user of an essential aspect of the news experience.

The second reason why the present invention finds particular application to the presentation of news is that news items are subject to constant updates while nevertheless retaining some measure of overlap over limited periods of time. Thus, a subscriber downloading the news at 06:00 and then again at 09:00 may expect many of the news items to remain unchanged. In the short period of time that elapses between successive views, there may be some new items of news that have not been presented previously and there may also be some updates to news items that were seen in the 06:00 transmission. However, current approaches to video content delivery constrain the subscriber to watching all items in the order that the content provider presents them.

Likewise, in the same way that a person who reads some newspaper articles while eating breakfast will not re-read the same articles when eating lunch, so too an Internet user does not wish to re-display the same headlines or expanded content twice. But according to regular usage, a user who accesses a news broadcast by typing a URL will be presented with at least substantially identical content. In other words, if the content has not been updated, the user will indeed see identical content. If it has been updated, the user will be able to display the new content second time around but will nevertheless be constrained also to see at least some of the original content. Furthermore, no clear distinction is made between previous content and new content because in many situations news content is subject to constant update and revision. This makes it onerous for the user to know where he left off from reviewing a previous article and if the content has been updated or changed. In practice, the user must scan the previous content to see where the subject matter appears unfamiliar.

The present application addresses the apparent conflict between several requirements, namely:

(i) A user wishes to receive video content of a type that may have been prior determined by the video host not to be of interest;

(ii) The user wishes to make a spontaneous decision whether to skip the video content or to receive expanded content relating thereto;

(iii) The user does not wish to receive repeated content during successive sessions;

(iv) The user does not want to receive extensive content that is of no interest; and

(v) The user wants to have a gapless and continuous viewing experience that is not marred by discontinuities between different video content.

Some of these requirements are at least to some extent mutually contradictory because once a user has indicated that a given type of media content (e.g. boxing) is of no interest, the decision as to what constitutes 'boxing' content lies in the hand of the host and therefore the user relinquishes control regarding whether specific content will be streamed. If it is not streamed because the host determines it to be unrequired, then the user is deprived of the choice to view it in depth. On the other hand, in the absence of such filtering by the host, the user will be bombarded with an abundance of content that may a priori be of no interest and whose streaming is therefore a burden on his time and on the resources of his media player. How to reconcile these requirements in a manner than creates a pleasant viewing experience without hiccups or discontinuities between successive media content is one object of the present invention.

It will be borne in mind that frequently and increasingly the media player of choice is that of a mobile platform, whose interface may be less comfortable for skipping unwanted content since their touch screens are typically navigated by means of the user's finger, which is too coarse to allow fine adjustment of the video player's timeline.

In one aspect there is provided a server system for streaming media content, the server comprising:

at least one processor,

memory coupled to the at least one processor, said memory storing a media file and a metadata file, the media file being playable on a media player to present a plurality of media content items in the media file as a continuous whole, at least some of the media content items being subdivided into a respective initial portion and a respective continuation portion, the metadata file comprising data that allows determination of respective start and end times of the initial portions according to a predetermined sequence, and

a respective communication module coupled to each processor for communicating with a user device and being configured to convey the metadata file to the user device and to stream sequentially initial portions in said sequence to the user device.

Typically the server system includes at least two servers, one for storing the media file and the other for storing the metadata file. However, both files may be stored on the same server and likewise portions of the media file may be distributed between different servers. Nevertheless, for the sake of simplicity of description, we will use the general term "server" in the singular, it being understood that no limitation is thereby implied.

In a typical scenario the media content server is connected to multiple user devices and streams successive items of video content in known manner. In one embodiment, by default successive lead items are conveyed one after another. The lead items may be news headlines, but they can be any media content that introduces a discrete video segment and together with the remainder of the video segment forms a continuous whole, while allowing a user viewing the lead item to make a spontaneous determination as to whether or not the whole video segment is of interest. The invention is not so much concerned with the application as with the concept and the detailed description will therefore relate to a news channel. An application/browser in the user device allows the user to request the complete item if the currently played lead item is of interest. Preferably, successive lead items are conveyed and buffered during the time interval that an earlier lead item is being played. This allows continuous, gapless streaming of lead items. For the sake of clarity, it should be noted that the term "streaming" relates to the user experience and is the result of displaying video content as it is conveyed in real time by the server. Likewise, the next story can be partially conveyed and buffered, so that it is available for immediate display if the user requests it. Such partial buffering will ensure that at the very least the lead item is immediately buffered and typically so, too, will the subsequent content since it follows directly from the lead item and forms a continuous whole therewith. However, these enhancements are not essential to the present invention.

Preferably, the server logs all items of video content streamed to each user device. If the server streams a news bulletin to a user device more than once and the later news bulletin includes items that were already streamed in an earlier bulletin, then those items that were streamed earlier will be omitted from the later news bulletin. This provides the user with a very simple and effective tool to obtain updates without the overhead of downloading the complete bulletin and sparing the viewer from presenting him the content already consumed.

In a typical embodiment of the invention, the user device displays media content received from the server but allows the user to signify interest or lack of interest in the currently displayed item, so as to control the streaming of subsequent media content. This may be done by displaying a visual marker for predetermined time period during which a media item is displayed and responsive to a user gesture at the user device and conveyed to the server during the predetermined time period that the visual marker is displayed, determining whether or not the user is interested in the displayed item. In one embodiment, the user device has a touchscreen, which the user can touch with a finger or stylus so as to relay to the host that the user wishes to see the expanded media item. Alternatively, the system can be configured to behave in the reverse manner whereby touching the marker indicates that streaming of the expanded item is not required, in which case the next lead item will be displayed. In one embodiment relating to news media, successive headlines constitute the lead items and are conveyed by the server and played on the user device. The user interacts with the user device in defined ways to display either successive headlines or to continue streaming and displaying the current item so as to display the expanded content associated that follows with the currently displayed headline. If the user device has a touchscreen, the user interaction can be any conventional gesture such as touching a visual marker or swiping the display screen in a predefined direction. For example, in a region where language is read left to right, swiping right can acts as a next item request and moves immediately to the next headline, while swiping left acts as a preceding item request that moves to the preceding headline. Swiping down or touching the marker can display the expanded item. At any time during the play of the expanded item, a user gesture may abort further display in favor of a subsequent lead item, which will otherwise be displayed upon completion of the expanded item.

In the specific case where all the lead items and the associated expanded content are spatially contiguous in the media file, it is sufficient to store in the metadata file only the start and end times of the lead items since the end time of a lead time will be the same as the start time of the associated expanded content and the end time of expanded content will be the same as the start time of the next lead item (or the end of the media file). But as noted previously, lead items and the associated expanded content need not be spatially contiguous. More generally, the lead items and the associated expanded content are distributed non-contiguously in the video file and there can be gaps between them. In this case, the metadata file stores the start and end times of the lead items as well as of the expanded content. The metadata file may omit items from the source video file provided by the publisher if the host so desires without the need to tamper in any way with the source file. In order to allow for gapless display between successive lead items, initial portions of successive lead items are preferably conveyed ahead of their being displayed and are buffered in the user device so as to be ready for immediate display if and when required. Likewise, in some embodiments, multiple instances of media player may be instantiated at the user device and each may be initial- ized for rendering a successive or previous lead item of the currently displayed item. Each instance of the media player corresponds to exactly one option of jumping from the current position. The appropriate instance of media player may then be activated to cause immediate display of the media content associated with a selected option. Multiple instances of media player may not be necessary where there are sufficient resources to allow for the gapless switching between various parts of video content.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to understand the invention and to see how it may be carried out in practice, embodiments will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which:

Fig. 1 shows pictorially a typical prior art arrangement for streaming content to a user;

Fig. 2 shows pictorially an arrangement according to an embodiment of the invention for streaming content to a user;

Figs. 3a and 3b are graphs showing respectively non-linear and linear streaming approaches;

Fig. 4 shows schematically a client-server system according to the invention for streaming media content;

Fig. 5 shows schematically details of the server;

Fig. 6 shows schematically details of the user device;

Fig. 7 shows schematically a media content set; and

Fig. 8 is a flowchart showing the principal operations carried out by the system.

DETAILED DESCRIPTION OF EMBODIMENTS

Fig. 4 shows schematically a system 10 comprising a media content server 11 that stores a database 12 or has access thereto either by a direct connection or via a communications network such as the Internet 13. The media content server 11 may be a single server that stores the database 12 or may include more than one distributed server. The media content server 11 is connected over a communications network such as the Internet 13 to multiple user devices 14, each associated with one or more users 17 who may be registered with the media content server 11. Fig. 5 shows schematically a detail of the media content host server 11, which comprises a transceiver 20 constituting a communication module that communicates with the user devices 14, typically over the Internet although other forms of communication may be employed. For the sake of illustration two different types of user device 14 are shown. One is a smartphone having a touchscreen 15 and the other is a conventional PC having a keyboard 16. The touchscreen 15 and the keyboard 16 constitute user interfaces that allow a user 17 to interact with the user device 14 and skip or display media content. But it will be appreciated that the user interface may be operated in other ways, for example using a mouse in combination with a regular PC or a stylus or pointer. The media content server 11 includes a processor 21 coupled to a memory 22, which is shown connected to the database 12. However, the database 12 may be distributed between one or more servers and for the purpose of the invention it is only necessary that the server 11 has read/write access to the database 12.

The memory 22 stores at least one media content set containing media content items described in detail below with reference to Fig. 5. The database 12 contains for each user device 14 or user 17 a respective record of media content items previously conveyed to the user device. The processor 21 is responsive to a stored program set for conveying to each user device, via the transceiver 20, a respective subset of media content items of the plurality of stored items from which some of the media content items are omitted. These include items already conveyed to and watched on the respective user device and optionally items that the server 11 has established are of no interest to the user. This may be established by monitoring which media content items streamed to the user device are skipped as will be explained shortly.

Fig. 6 shows schematically a detail of the user device 14, comprising a transceiver 30 constituting a communication module that communicates with the server 11, typically over the Internet although other forms of communication may be employed. The user device 14 includes a processor 31 coupled to a memory 32, a user interface 33 and a media player 35. The user interface 33 may include the touchscreen 15 or the keyboard 16 shown in Fig. 1 and is used to control the manner in which media content is streamed to the user device and displayed thereby and to send feedback to the server 11 , which may be employed to provide customized streaming to the user device. The memory is partitioned into a number of discrete buffers #1, #2, #3 and #4 as well as a transition buffer shown respectively as 32a, 32b, 32c, 32d and 3e whose significance will become apparent in the following description of an application. The media player 35 is shown functionally but is typically implemented in software stored in the memory 32 and executed by the processor 31 and displays visual media content on a display 36, which may be the touchscreen of a smartphone although the invention is not restricted for use with touchscreens, although this is probably the preferred method of use.

Fig. 7 shows schematically a media content set 40 relating to a news bulletin comprising a plurality of media content items arranged as six news stories comprising six headlines 41a-41f and six expanded news content 42a-42f each relating to the respective headlines. Each expanded item follows continuously from the associated lead item, so that if the media content set is streamed and displayed in its entirety, the six news stories follow gaplessly one after the other. For ease of description the items are labelled so as to depict how they might be presented in a typical news bulletin. Thus, the bulletin might start with an item of important local interest followed by an item of important global interest. After each headline 41a-41f is streamed to the user device there may follow a corresponding expanded item 42a-42f. The manner in which the news items are organized is not important since the start times of each item are stored in a metadata file that is separate from the media content. The start times of successive items define a sequence, which need not be arranged in the same order in which the items appear in the media content set 40. However, for the sake of description and clarity, it will be assumed that the news items follow in logical order, whereby the start times of the lead items and associated expanded content are in increasing order. In one particular application to which the present invention may be usefully applied, the server does not stream both headlines and full expanded news content unless specifically requested to do so by the user. Thus, by default only the headlines are streamed to the user device. The dashed line joining the media items shows the timeline, it being seen that for the n^th story, the corresponding lead item starts at a time t_no and the corresponding expanded item starts at a time t_nj. This does not require that the associated expanded content commences in the video file where the lead item ends, i.e. that the two are spatially contiguous in the video file. They can be spatially separated but they will still be streamed as a continuous whole because the user device uses the metadata to stream the corresponding video items from the server one after the other to maintain temporal continuity. In some embodiments, the media file is a unitary file each of whose constituent media content items is accessed from a single URL, while in other embodiments, at least two segments of the media file are accessed from different URLs.

During the act of viewing a headline, the user may decide that this item warrants a more detailed presentation. In such case, the user may use the user interface 33 in order to send information to the server identifying that he or she wants to receive the expanded item. In one embodiment, the user device includes a touchscreen and determines from the metadata file whether a headline has expanded content. If so, the user device displays an icon, which when touched by the user causes the user device to stream the expanded item from the server after completing streaming the headline. The processor 31 in the user device 14 is responsive to a stored program for displaying the icon and for allowing the user to indicate that the current headline is to be followed by the respective expanded item. The program also allows the user to skip media items using the user interface 33. This can be done by swiping the currently displayed item so as to advance to the next item; or it can be done by displaying an icon which when touched by the user advances to the next item. When a user skips an item this may be communicated to the server 11 and stored in the respective record of the database 12. The server 11 may thereby infer that this item is of no particular interest to the user. For example, if the user skips the fifth item relating to open golf, the server might infer that the user is not interested in golf. The server 11 may be programmed to analyze the data statistically in order to determine whether a skipped item is statistically significant or just a random occurrence, possibly due to lack of time or even unintended action of the user interface. For example, if every time a golf item is streamed the user skips to the next item, the server will conclude that the user is not interested in golf; but this does not necessarily imply that he is not interested in sport since he may evince an interest in baseball or tennis. The manner in which a determination is made by the server 11 to omit certain types of media content from subsequent transmissions to a given user device is not itself important other than to note that it is always based on user feedback via the user interface 33. Thus, the server is adapted to analyze information relating to activity received from the user device when watching and terminating media content, the information permitting the server to determine an identity of the media content and durations of watching thereof, and to update the respective database record accordingly. These operations are summarized in a flow chart shown in Fig. 7.

In order to allow a user to gaplessly skip from one headline to another, during the time that the first headline 41a is being streamed from the media content server 11 to the user device where it is buffered in the first buffer 32a and played by the user 17, in parallel with the first headline 41a at least one and preferably more of the subsequent headlines 41b, 41c etc. are also downloaded and buffered.

Within the context of the invention and the appended claims, it is to be understood that skipping a media item is not restricted to a wholesale rejection thereof. For example, a user may start watching a media item, such as a headline or even an expanded item and decide to abort the item prior to its proper termination. The present invention is not concerned with how the server determines whether intermediate termination is to be construed as a lack of interest in the displayed item or media content type or is merely the result of the user having seen sufficient of the displayed item to serve his or her immediate needs. One simple way of achieving this is to base user interest or the lack thereof on the fraction or duration of content actually viewed. For example, if the user watches more than 25% or 60 seconds of the media content before terminating, this may signal that the reason for termination is lack of time rather than lack of interest; while if the media content is terminated after watching less than 25% or 30 seconds this may be used to signal a lack of interest. Even when a lack of interest in a specific media item is determined, this does not necessarily imply that the user is not interested in the underlying media content type. As noted above, such information may be established by the server by subjecting repeated user interactions to statistical analysis. It could also be based on the display and user selection of a different icon or of some other user interaction such as swiping downwards or pinching the touchscreen of the user device or even providing a setup program that allows the user to disable specified content types via a keyboard or pointing device.

It should also be noted that for the purpose of the above description no distinction is made between "user device" and "user". In the case that only a single user ever uses a user device the distinction is in any case moot. But the invention allows for multiple users to be registered with the same device and for a number of devices to be registered in the name of the same user. In this case, the user may be identified to the server, which will then create and maintain in the database a record pertaining to the user. The content subsequently streamed to the user device will then depend on the identity of the user, which may be established either manually via the user interface 33 or using a camera (not shown) built-in to the user device or via whatever other means. In the case that the database creates user-specific records, these will typically be used to supersede device-specific records so that users who are registered with the server will receive customized media regardless of which device they use to connect to the server.

The description so far has related to user-preferences which result in the server omitting media content of specified type from subsequent sessions. For example, by way of example only, the media content types associated with Fig. 4 may be "local", "finance", "foreign", "celebrities", "sport" and "weather". In practice, media content may be tagged with multiple types in order to provide finer resolution of which content is of interest and which may be suppressed in order to avoid streaming redundant data to the user device.

As noted above, additional savings can be achieved by omitting from a subsequent session those media content items that were streamed in an earlier session. Thus, reverting to Fig. 5 consider the news that Bob Dylan was awarded the Nobel Prize for Literature in 2016. Immediately following the first announcement, Bob Dylan was unavailable for comment and for over a week there were substantially no new items of interest to report. Eventually he responded and expressed his delight. Subsequent to this there were reports of his preparing his Nobel speech and then later that he would not in fact be attending after all. This is an example of updatable news items and is of course standard fare in the news industry. After a user streamed the media content item relating to Bob Dylan following the first announcement, there was no need to stream the same item during successive sessions to the same user or device until eventually there was something new to report. In some embodiments, the database 12 maintains a current record for each user or user device of which media content items have been streamed and omits the same items from subsequent sessions. On the other hand, when new media content becomes available it does not necessarily replace the previous item relating to the same subject. It may instead be used to create an updated item that both announces the award to Bob Dylan and informs of his reaction. A user who has already streamed the first announcement will then be streamed only the new item relating to Bob Dylan's reaction. But a different user who has not yet streamed the first announcement will receive the updated item that includes both the announcement and the reaction.

The above embodiments relate to news bulletins because news content is constantly updated and successive bulletins therefore contain many identical segments, with later bulletins containing relatively sparse new content. Furthermore, news content is targeted at a broad audience and therefore includes many types of video content, not all of which may be of interest to any one specific user. But it is to be understood that the invention is not limited to news media and will find broader application whenever repetitious streaming of media content containing multiple items or segments that are subject to incremental updates over time is to be avoided. It will also find application where redundant streaming of media content containing multiple items or segments some of which are known to be of no interest to the user is to be avoided.

It should also be noted that features that are described with reference to one or more embodiments are described by way of example rather than by way of limitation to those embodiments. Thus, unless stated otherwise or unless particular combinations are clearly inadmissible, optional features that are described with reference to only some embodiments are assumed to be likewise applicable to all other embodiments also.

It will also be understood that the user device according to the invention may be a suitably programmed computer. In this context, "computer" designates any device having a processor, communications interface and memory. Likewise, the invention contemplates a computer program being readable by a computer for executing the method of the invention. The invention further contemplates a machine -readable memory tangibly embodying a program of instructions executable by the machine for executing the method of the invention.

Claims

CLAIMS:

1. A server system for streaming media content, the server system comprising: at least one processor,

storage coupled to the at least one processor, said storage storing a media file and a metadata file, the media file being playable on a media player to present a plurality of media content items in the media file as a continuous whole, at least some of the media content items being subdivided into a respective initial portion and a respective continuation portion, the metadata file comprising data that allows determination of respective start and end times of said portions according to a predetermined sequence, and

at least one communication module coupled to the at least one processor for communicating with a user device and being controlled by the at least one processor to convey the metadata file to the user device and to stream sequentially initial portions in said sequence to the user device.

2. The server system according to claim 1, being responsive to a continuation request received from the user device for streaming to the user device the respective continuation portion associated with a currently displayed initial portion.

3. The server system according to claim 1 or 2, being responsive to a next item request received from the user device for streaming to the user device the respective initial portion of the media content item that follows a currently displayed initial portion in said sequence.

4. The server system according to any one of claims 1 to 3, being responsive to a preceding item request received from the user device for streaming to the user device the respective initial portion of the media content item that precedes a currently displayed initial portion in said sequence.

5. The server according to any one of claims 1 to 4, wherein the media file is a unitary file each of whose constituent media content items is accessed from a single URL.

6. The server system according to any one of claims 1 to 4, wherein at least two portions of the media file are accessed from different URLs.

7. The server system according to any one of the preceding claims, being adapted to stream to the user device multiple portions of the media content set simultaneously.

8. The server system according to any one of the preceding claims, further including a database containing at least one record for the user/user device that identifies media items that have been previously streamed and rendered by the user/user device, wherein:

the system is adapted to omit said media items during a subsequent session in respect of said user/user device.

9. The server system according to any one of the preceding claims, further including a database containing at least one record for the user/user device that identifies usage of content specific to said user/user device, and

wherein the server system is adapted to analyze information relating to activity received from the user device when rendering media content, said information permitting the server to determine characteristics of the media content that are of interest to the user, and to update the respective database record accordingly.

10. The server system according to any one of the preceding claims, wherein:

the user/user device is registered with the server system,

the database contains records for the registered user/user device one or more media content characteristics that are of interest to the user/user device, and

the system is adapted to stream to the user/user device only media content items corresponding to said characteristics.

11. The server system according to any one of the preceding claims, wherein:

more than one user is registered with the server system,

the database records for each registered user one or more media content types that are of interest to the user, and

the system is adapted to identify the user and to stream to the user device only media content items corresponding to said media content types.

12. A user device comprising:

a processor,

a communication module coupled to the processor for downloading a metadata file from a host server coupled to the communication module,

a storage for storing the metadata file,

said communication module being responsive to the processor for streaming media from a media content server to the user device to present a plurality of media content items as a continuous whole, at least some of the media content items being subdivided into a respective initial portion and a respective continuation portion, and a user interface coupled to the processor for allowing a user to convey requests to the media content server during the rendering of media content, said requests including timing data determined from the metadata file.

13. The user device according to claim 12, wherein the processor is adapted to stream multiple segments of the media file simultaneously.

14. The user device according to claim 12 or 13, wherein the processor is responsive to a continuation request received from the user interface for streaming the respective continuation portion associated with a currently displayed initial portion.

15. The user device according to any one of claims 12 to 14, wherein the processor is responsive to a next item request received from the user interface for streaming the respective initial portion of the media content item that follows a currently displayed initial portion in said sequence.

16. The user device according to any one of claims 12 to 15, wherein the processor is responsive to a preceding item request received from the user interface for streaming the respective initial portion of the media content item that precedes a currently displayed initial portion in said sequence.

17. The user device according to any one of claims 12 to 16, wherein the user interface includes a touchscreen and the continuation request, the next item request and the preceding item request are respective user gestures.

18. The user device according to any one of claims 12 to 17, wherein the communi- cation module conveys an identity of the user to the server and receives from the media content server only media content items corresponding to the characteristics that have been determined by the media content server to be of interest to the user.

19. The user device according to any one of claims 12 to 18, wherein the processor is configured to instantiate at least two instances of media player each of which corresponds to exactly one option of jumping from a current position.

20. The user device according to any one of claims 12 to 19, wherein the processor is responsive to the metadata for suppressing or hiding a seek bar from the media player while headlines are being watched.

21. The user device according to claim 20, wherein the processor is responsive to the metadata for displaying the seek bar while expanded content is being watched.

22. A server system for streaming media content, the server system comprising: at least one processor,

storage coupled to the at least one processor, said storage storing a media file and a metadata file, the media file being playable on a media player to present a plurality of media content items, the metadata file comprising timing data that allows determination of respective start and end times of the media content items according to a predetermined sequence, and

a respective communication module coupled to each processor for communicating with a user device and being configured to convey the metadata file to the user device, to receive from the user device timing data of requested media content items extracted from the metadata file and to stream the requested media content items to the user device.

23. A user device comprising:

a processor,

a communication module coupled to the processor for communicating with a server system for streaming media to the user device,

a storage coupled to the processor for receiving from the server system a metadata file comprising timing data that allows determination of respective start and end times of media content items in a media file according to a predetermined sequence and for buffering portions of the media file playable on a media player coupled to said storage to present a plurality of media content items in said sequence, and a user interface coupled to the processor for allowing a user to convey requests to the server system during the display of media content, said requests including timing data determined from the metadata file relating to preceding or succeeding media content items in said sequence that are to be streamed from the server system.

24. The user device according to claim 23, wherein the processor is responsive to commands entered via the user interface to access the metadata file, extract therefrom data representative of a start and end time corresponding to a next or preceding lead item or an expanded item relating to a currently displayed item, to convey to the server system a streaming request that includes said start time, to stream the requested media item from the server system and to feed it to the media player until the end time is reached.

25. The user device according to claim 23 or 24, wherein the processor is configured to instantiate at least two instances of media player each of which corresponds to exactly one option of jumping from a current position in the media file to a position corresponding to a preceding or succeeding media content item in said sequence.

26. A method for streaming on-demand video, including:

downloading a metadata file comprising timing data that allows determination of respective start and end times of media content items in a media file according to a predetermined sequence,

receiving user commands for navigating linearly between sequential media items arranged in a predetermined temporal order and for either skipping or streaming items such that each command is adapted to extract timing data associated with a selected media item being a successive or preceding media item or an expanded media item relating to a currently watched media item and to use the timing data to stream the selected media item;

directing a request to a server for streaming the selected media; and

streaming the selected media to a media player.

27. A computer program product comprising a machine readable memory storing program code, which when run on a computer executes the method of claim 26.