Video encoding can be a complex and complicated space. I often get requests from customers asking for ways to transcode formats that are used primarily in broadcast studios, or in the movie industry. Accompanying these formats are also complicated ‘recipes’ to process the videos that requires custom logic to make decisions about what encoding steps are required.

This week, I am excited to introduce a new technology in Microsoft Azure Media Services that allows our more experienced customers to solve such workflow problems. This premium encoder offers superior quality and flexibility suited for broadcast industry/professional media transcodes. This includes automated decision-making logic that can adapt to a variety of input and output file formats such as MXF, support for additional input and output codecs such as DNxHD, support for higher output resolutions – up to 4K in AVC, support for closed captions,  and much more.

In this blog, I will provide you an overview of the new capabilities of the premium encoder. Please click here to find out more about pricing and visit our documentation center to learn more details about how to use this encoder.

Capabilities of the Premium Encoder

In the following sections, I will share some of the highlights about the new encoder.

Automatic Decision-Making Logic

Our “Azure Media Encoder” is based on a desktop encoder (Expression Encoder 4). In this application, you had to manually select and import each media file, and choose encoding settings after the file has been inspected. It works great – if you have the time to first determine what the properties of the input video file are: 1080p vs 720p vs SD, or 5.1 audio vs. stereo audio. I have been asked by many customers for a way to define one encoding preset that works for any source, 1080p or 720p or SD, and ensures that output is no bigger than the input (either in size or in video resolution). That is, if you send in an SD source to the encoder with that preset, the output adaptive streams are at most at SD resolution. Else if you send in a 720p source with the same preset, again the output streams are at most 720p, but with higher bitrates for the HD layers.

With the premium encoder, you can tell it to use such decision-making logic – see the diagram below, which represents one encoding preset, aka workflow.

With the “Azure Media Encoder”, we provided you with some default configurations or encode settings. Similarly, for this encoder, we will be providing encoder configurations for most commonly used scenarios. The encoding configurations for the premium encoder are referred to as workflows. While some default workflows for most commonly used scenarios are provided for you, you can create custom workflows using the Workflow Designer tool (details below).

Support for Additional Formats and Codecs

With the premium encoder, you get support for additional formats and codecs that are popular with the television broadcast industry. You can import or export MXF files, which codecs such as DNxHD, DVCPro/DVCProHD, JPEG2000, and much more. You can visit our documentation center to learn more details.

Support for Closed Captioning

In this initial release, the premium encoder supports closed captioning when it is interleaved into the source or input media file: essentially, CEA-608/CEA-708 carried as user data (SEI messages of H.264 elementary streams, ATSC/53, SCTE20) or carried as ancillary data in MXF/GXF files. The encoder can then either pass through the captioning data to the output video file(s), or save the data to a side-car file.

Audio and Video Processing Capabilities

There are way too many toys built into the premium encoder to describe in depth here. I’ll call out some of my favorites:

• Broadcast quality de-interlacing with inverse telecine – for example, it can detect whether the input has a cadence of 2:2, 3:2, 2:3:3:2, 2:2:2:4, 3:2:3:2:2, 5:5, 6:4, 8:7 and so on.
• Dolby® Professional Loudness Correction with Dialogue Intelligence™ – see this for more information
• Ability to detect and remove black borders (pillarboxes/letterboxes), so that only the active picture area is transcoded. It can also insert black borders, if needed
• Ability to add overlays over the video, including graphics and timecode
• Support for sources with multiple language audio tracks

When to use the Premium Encoder

You can use the flow chart below to decide whether your transcoding needs require the use of the premium encoder or the standard Azure Media Encoder. For simplicity, the capabilities of the latter encoder are not documented here. See here for details on the supported file formats and audio/video codecs.

Decision tree to determine which encoder to use

Workflow Designer

From the previous section on the automatic decision-making logic, it should be clear that the new encoder is very powerful, and can be configured to run very complex workflows. Such configuration is made possible via a desktop application called the “Workflow Designer”. Below is a screenshot of a sample workflow.

Sample Workflow for Premium Encoder

Shortly after the preview release of this encoder, you will have access to this standalone desktop tool that can be used to modify the default workflows, and create your own workflows. To get a copy of this Workflow Designer tool, please contact mepd@microsoft.com.