The captions you added in the previous module are an example of side-car metadata. That is, metadata that is stored separately from the video. Videos can also have embedded metadata that is stored in the video package itself. In this module, we will look at some common examples of embedded metadata including Ad markers, embedded captions, and multi-language audio tracks. We will use Ad markers as an example to show how to work with embedded metadata in the video conversion workflow.
The video you have been working with contains 2 language tracks and SCTE35 Ad Markers. Ad markers show where ads should be located in a video. Ad marker blanking inserts a visible slate into the slot that is reserved for an Ad. Ad can be inserted on playout using a packager such as AWS Elemental MediaPackage.
You need to have access to MediaConvert and S3 to complete this module.
You need the following resources created in module 1:
- MediaConvertRole - the role created to give permission for MediaConvert to access resources in your account.
- MediaBucket - the bucket created to store outputs from MediaConvert.
- MediaConvert job from module 2 or later - We will start with this job and modify it in this module.
If you used CloudFormation to configure resources, you will find the values names of MediaConvertRole and MediaBucket in the Outputs of the Stack.
Create a duplicate of the job you created in the last module. Modify the job to detect add markers and replace any embedded Ads with a slate image. The new job will have the following structure:
- Open the MediaConvert console for the region you are completing the lab in (US-West-Oregon). https://us-west-2.console.aws.amazon.com/mediaconvert/home?region=us-west-2#/welcome
- Select Jobs from the side bar menu.
- Find the job you created in the last module and click on the Job Id link to open the Job details page.
- Select Duplicate
- Make sure the input video path for Input 1 is set to:
s3://rodeolabz-us-west-2/vodconsole/VANLIFE.m2ts. This video has been created with SCTE-35 Ad Markers inserted.
- Scroll down to the bottom of the MediaConvert side bar menu and select Settings from the Job settings section.
- Scroll down to the Global processors panel and toggle the Ad avail blanking switch to the on position.
- Enter
s3://rodeolabz-us-west-2/vodconsole/SLATE.pngin the Blanking image box.
- Scroll to the bottom of the page and select Create
- Wait for the job to complete. Monitor the status of the job by refreshing the Job detail page.
Ad markers will now show up in all of the video outputs.
To play the videos, you will use the S3 HTTPS resource Link on the videos S3 object Overview page.
The MP4 output is located in your ouput s3 bucket in the object: s3://YOUR-MediaBucket/assets/VANLIFE/MP4/VANLIFE.mp4
You can play the MP4 using:
- Chrome by clicking on the Link for the object.
- JW Player Stream Tester by copying the link for the object and inputing it to the player. https://developer.jwplayer.com/tools/stream-tester/
The HLS manifest file is located in your ouput s3 bucket in the object: s3://YOUR-MediaBucket/assets/VANLIFE/HLS/VANLIFE.m3u8
You can play the HLS using:
- Safari browser by clicking on the Link for the object.
- JW Player Stream Tester - by copying the link for the object and inputing it to the player. https://developer.jwplayer.com/tools/stream-tester/
The MP4, HLS and Thumbnails will have one 10 second Ad marker slate image inserted starting 6 seconds into the video.
In this module you learned about embedded metadata in file based video conversion workflows. In the next module, you will switch gears from the functionality of mediaconvert and focus on automation by creating a lambda function that starts a new MediaConvert job whenever a new file is written to an S3 bucket.
Next module: Automating Jobs with Lambda