6 features of Flussonic you might not know about
The Flussonic family of products is one of the most intensively developing technologies in OTT media market. New features and improvements get into production with each monthly update. Some of them have been added by the Flussonic team as part of the software development roadmap, others – inspired by client requests and market trends. We always try to provide detailed release notes, but sometimes the most interesting features elude the attention of the users. So we’re going to review them once more in this special post.
Table of contents:
- Subtitles OCR
- Transcoder
- Smooth Source Switching
- Low latency playback
- EPG
- Watcher SDK
Subtitles OCR
Such a step is usually made in DVB which is an abbreviation for Digital Video Broadcast because a subtitle picture is created as a “picture” in the MPEG-TS stream from a satellite to avoid problems with fonts and text drawing on different devices. After the version 4.7.2, the video server type Flussonic could identify the digital broadcast subtitles which use WebVTT as its base and process it using a standard unicode text form, that is UTF-8, to limit the costs of constructing the video stream dramatically.
WebVTT (Web Video Text Tracks Format) is the most common-yet-powerful subtitle format which is supported by web browsers and gives other stuff too. In addition to the use of CSS, people might fool around with the font and spice up with visual effects in order to get an individual approach.
Then why do we not use SRT/subtitles? VTT is certainly more advanced here in the functionality, which has some many very good features for complex multimedia localization, including metadata parameters and formatting options.
Transcoder
Transcoding is necessary if you want to:
- create a multi-bitrate stream
- change parameters of video — the codec and the bitrate of the stream, the size of the image
- overlay a logo on top of a video stream.
Flussonic Media Server has its transcoder incorporated. It embodies transcoding by using GPU or CPU.
Transcoder module works together with any source file supported by Flussonic Media Server.
HLS is partially supported — certain sources can not be transcoded. It is suggested to evaluate each HLS source by hands to check whether it works correctly after transcoding.
For hardware-accelerated transcoding, Flussonic can use Intel and Nvidia solutions. When transcoding with Nvidia NVENC, streams with a 10-bit color depth are supported.
Smooth Source Switching
Under different cases of video source missing or even shutting down. To prevent a situation where there are no videos available, it is important to provide alternative sources for the stream, which will be used for broadcasting when the main source is down until the restoration of that main stream. The automatics source switching technology comes in handy at Flussonic.
The Flussonic’s transcoder does the work of handling transformation of the properties of a video stream just fine. While switching a source, the Flussonic’s transcoder does not lose a single frame if even the media information has changed. The most commonly experienced situation is the alteration of the size of the photo in video from the new source to which the stream has recently switched. One example is using different width and height ad when a file or file of different width and height is inserted into a stream, or when transcoding a server-side playlist consisting of files and streams of various resolutions. Flicker and incoherency can appear in the continuous stream due to the dropped frames.
If you transcode by using the CPU or Nvidia NVENC, seamless source switching is the default behavior. Starting from Flussonic 20.07, our CPU transcoder is able not to lose any frames at the change of a stream’s media information.
Low latency playback
The reference to low latency is in OTT video services is just one of the emerging trends, and this phenomenon will be one of the most heated topics for many years to come. To be a winner of this competition, we implement 3 new technologies which, while tested, deliver the same upshot in around 300ms on a stand and a wrench.
Initially we designed a multidirectional player with low-load player. Absolutely, we blog our code development /publicize it on GitHub for you to read, if you like it, enjoy! Its all about “Zero-Latency-Player”. Very easy to remember and it does not make us distant from our modesty. The player offers a low-delay HTML5 playback without the necessity of the old nasty Flash (it’s 2018). The player is working when WebRTC can only fail. As before, we create a thing of our embed.html which you should dump in the front-end project. But now you get the opportunity to use javascript module.
Second, we mastered WebRTC, which means that Flussonic Media Server can do both - receive and send WebRTC video fast and smooth. It’s a really unique feature - many media software enthusiasts challenged it’s complications but only few managed to create stable solutions that use the full potential of WebRTC.
And finally, we’ve made fine tuning of Flussonic DVR tools to optimize time spent for internal indexe updates. It seriously minimized storage load and, as a result, influenced the overall playback latency.
MSE-player GitHub page: https://github.com/flussonic/mse-player
EPG (Electronic Program Guide)
With the last update we presented new tools to extract program schedule from the stream, for example if satellite TV transmits EPG data together with broadcasts in MPEG-TS. Now it’s up to admin to decide whether he wants to use our free embedded solution or to buy data from third-party services. Both approaches have their own pros and cons. As Flussonic only extracts a schedule from the main source, it keeps the data as-is, which means that consistency and accuracy of the schedule depends on content provider, without double check from Flussonic.
However, Flussonic feature is a helpful feature if it bothers us of the risks that we similarly know. Being the only core component, the FlashLocator of Flussonic holds the archive video data as a non-stopping timeline, this is definition of continuous timeline data. In the case there a wrong time in a schedule as lifetime displays a cue at the actual time of transmission not the timeline of the roundup of news in a file.
Flussonic exports EPG into two formats: Xmltv - as it permits text broadcasts describing TV programs, which is predominantly used in IPTV middleware, and JSON - for web sites integration for displaying the program on web pages.
Official documentation: https://flussonic.com/doc/iptv/export-of-epg-from-mpeg-ts-streams
Watcher SDK
Finally let’s talk about the feature we’ve never publicly announced yet. Of course, you’ve heard a lot about Flussonic APIs of Media Server or Watcher. But we went further and decided to create a fully functional SDK to embed Flussonic technologies into custom mobile applications.
A significant number of our customers are providers of end-user services: IPTV, VOD and video surveillance. Their success depends on the quality of video delivery, including low latency, fine picture and accessibility of video content. It takes serious effort and expenses to create a competitive tool for video playback, that’s why the Watcher SDK is extremely helpful.
The fact that the Watcher SDK is a Flussonic MSE-player which allows developers embed it into their mobile application for Android and iOS they will get the long list of positive features of the MSE-player. The app is capable of a video playback that is possible from the rest of the application which allows the user to access the video files and streams via the Flussonic API.
The use of Watcher SDK allows the creation of a brand new application with a top-quality media player, individual appearance, and customized settings based on certain business requirements at once.Cool, yeah? We have already included this killer feature in the production phase.