Mobile video encoding

As TV service providers try to keep up with the surge of mobile video consumption and more connected TVs and tablets coming to market, the need for cost-effective and secure encoding services that provide quality of experience becomes increasingly important. And, although TV service providers understand mobile is important, are they really ready to deliver these services to consumers? Are they prepared to roll out services to stay competitive, but also provide a high-quality experience that protects the quality of their brand?

According to a recent report from In-Stat, worldwide revenue for multiformat transcoders will surpass $460 million in 2015. The reason behind this growth is that due to varying OS and screen size, every new device requires service providers to individually encode video for each model or device family. Furthermore, with the adoption of adaptive streaming, there are now multiple variants of each device. For cable operators, TV service providers and movie studios — or anyone providing premium content to a subscriber base — the need for better encoding is even more critical as the mobile video adoption curve climbs upward and consumers demand a pristine viewing experience on their smart-phone and tablet devices. Premium content can be defined as a television show, a movie or a live sporting event — essentially, anything that consumers will pay to watch. Today, the most common means of watching this content is through cable subscriptions. However, as mobile video becomes mainstream, consumers who wish to have access to their cable service across multiple screens will have the choice to either purchase a premium bundle or pay more for a basic package in order to view this same content on a mobile device.

Premium content requires a precise set of specifications, all of which are dependent on factors such as network capacity, carrier requirements, target devices and diversity of source formats. The encoding process follows a four-step process: collect, transcode, encrypt and publish. Although this process is not new, the specifications associated with it continue to evolve due to device capabilities, network evolution, and enhanced compression and delivery technologies. For a bit of perspective, consider this: On average, a 22-minute episode of premium content requires between 20 and 25 output files to accommodate varying devices, screen sizes and (where appropriate) adaptive streams. This is a challenge for content providers across the board. According to an article from Leslie Ellis' “Translation Please” blog, HBO executive Diane Tryneski states, “Since 2006, the number of video assets HBO creates every month went from 500 to 60,000.” This story is a familiar one across the industry. As long as new devices are introduced to the market, this number will continue to climb, and content providers will be faced with the challenge of pushing more content in more formats to more devices and increasing network capacity.

As one can imagine, the encoding process can become a time-intensive and costly initiative for content providers to handle on their own. Although many have tried, it requires a lot of resources, which means many service providers are now turning to third-party specialists to help streamline the process and make it more cost-effective, while leveraging economies of scale. Third-party specialists also are better aligned to keep on top of new developments, standards and innovations. In order to provide best-in-class live-TV and VOD services for mobile devices, it's important to consider the points below.

Ensuring high-quality VOD

Aside from the content selection, video playback and quality are the most important parts of the mobile video experience from a consumer's perspective. Thus, it is critical that cable, TV service and content providers have stringent benchmarks in place when testing the quality and delivery of premium content. The first stage of analysis often happens during the ingestion process. When working with a third-party encoding vendor, service providers should make sure that they are receiving optimal source files and accurate metadata for the type of VOD content they are providing. They should understand the preferred levels of interlacing, aspect ratios, resolution, codecs, frame rates and bit rates in order to ensure that they are starting with the best source file. From there, the files should be tested and checked to confirm they meet agreed-upon standards. After the source input is successfully validated, a series of pre-processing steps take place. These steps include video equalization, de-interlacing the file, audio boosting and noise reduction. The encoding process may then begin. If any of these steps are skipped prior to the encoding process, it could severely impact the delivery and quality of the content.

From here, many vendors take a customized approach to the encoding process. The most challenging part of this is determining how many encodes are required to meet the specifications of the overall service. It is important to evaluate key criteria such as network and content types, media player preferences, security, and device capabilities and, in some cases, carrier restrictions. By separating the audio and video layers during performance testing, one can ensure a robust and error-free encode and shape the encoding profile for various device/service combinations. However, it is likely that these encoding profiles will need to be refined as new services and devices come to market.

After each encode is created, each file should be analyzed using automated QC tools to ensure that it meets the quality expectations of the service. If errors are detected during this stage, then manual validation is often required to determine severity levels and next steps. In most cases, a faulty encode will need to be re-encoded from scratch. It should also be standard practice to manually sample video on actual devices to ensure the final product is of the best possible quality and free of any potential errors.

Live TV — no room for error

It goes without saying that with live TV, there is little opportunity to refine and manipulate the content as it makes its way to a variety of IP-connected devices. That does not mean the quality has to suffer. Content and service providers should feel confident that their services have undergone rigorous testing and monitoring to provide a quality experience for consumers. This includes ensuring that there is a high-quality input feed, that adaptive streaming protocols are supported between various networks, and that the encoding service provider has configured specific profiles for all content. Most importantly, encoding service providers must make sure that any issues in the redundant infrastructure are addressed to avoid the potential of having a negative impact on the service.

The architecture at the core of live TV streaming services should be taken seriously. Whether one is trying to deliver the content via satellite, SDI, ASI or IP speeds, it is important to establish a connection with service provider headends.

Adaptive streaming

Due to its consistency and significantly improved quality, adaptive streaming has taken over as the trusted model for streaming live video. However, despite the fact that Real Time Streaming Protocol (RTSP) is no longer the protocol of choice for streaming, it is important that encoding service providers offer support for legacy devices that still use this protocol and maintain capabilities for distributing the content.

Adaptive streaming comes with some challenges: Variables such as bit rate and resolution must be carefully analyzed and tested so that the media player can quickly transition to new versions without compromising the integrity of the picture. Adaptive Streaming now powers many of the early TV Everywhere efforts from cable service providers such as Comcast and Time Warner. These services are helping to drive the adoption and pervasiveness of Adaptive Streaming by making it easier to get various types of content to different devices over several types of network connections. The ability to push higher-quality versions will really come to bear as Long-term Evolution (LTE) networks and devices get rolled out.

The next encoding frontier

A plethora of options exists for determining how to collect, transcode, encrypt and deliver premium mobile video. One thing remains constant: The encoding process is a critical step in that life cycle. Spending too much on encoding processes or encoding inaccurately can negate a tremendous opportunity to drive additional revenue or value and keep subscribers happy. It is important to find an encoding service provider that will deliver a high-quality finished product and document the process from start to finish in order to establish timeliness in a cost-effective way.

Derek Bell is senior director of product management for QuickPlay Media.