Step 5: Output (CMAF) - How do I configure CMAF Output for my Service using the new UI?
Learn how to configure CMAF as the Output source using the New Services UI.
In this step, you will configure where and how your processed captions, subtitles, and dubbed audio are delivered. For a list of all steps see our Service Wizard Steps below.

Stream Type: Select CMAF as your stream type.
Delivery Method: Choose how your output stream will be delivered. For CMAF output, only PUSH is available.
CMAF PUSH
Use this option when SyncWords will push the output stream with processed captions, subtitles, and dubbed audio to your specified CMAF destination using the Common Media Application Format (CMAF). This setup is ideal for low-latency streaming workflows and is commonly used with AWS MediaPackage v2 and other modern CMAF ingest platforms.

Destination: Select the downstream service that will handle the CMAF ingest stream. Available options include:
- MediaPackage v2 (recommended) - Optimized output settings for AWS MediaPackage v2
- Ateme Encoder - Optimized output settings for Ateme encoding platforms
- HTTP Endpoint - Standard CMAF output for generic HTTP ingest endpoints
Ingest Endpoint: Enter the CMAF Ingest Endpoint URL where the Service should send the output stream. Copy the CMAF Ingest Endpoint URL from your AWS MediaPackage v2 channel or other CMAF packager and paste it here.
IMPORTANT: If you are publishing to MediaPackage v2, you must add a policy to your channel. There are two methods available to do that. See articles below for additional help - Method 1 or Method 2
Mode: Choose how the CMAF stream will be delivered. You can select between RELAY and APPEND.
- In Relay mode, SyncWords will push the entire input CMAF stream to the CMAF endpoint along with any subtitles and dubbed audio tracks.
- In Append mode, SyncWords will only contribute subtitles and dubbed audio tracks.
Buffer: Set the buffer duration in seconds to manage latency and ensure stable output transmission. Latency allows perfected timing of captions, subtitles, and dubbing with the audio in the stream.
- The default 30-second buffer is recommended when using translated subtitles
- A 60-second buffer is recommended when enabling dubs, as audio synthesis requires additional processing time
- For low-latency delivery of same-language captions, use a value greater than your HLS segment size or a minimum 5-second buffer
NEXT STEP: Once you’ve entered the required information, click "Continue" to move to Step 6: Preview where you will preview & create your Service.
Navigation Options
At the bottom of the screen, you will see the following actions:
- Cancel - Cancels the creation of your Service and exits the wizard.
- Reset - Clears all fields and settings in the Wizard so you can start over.
- Step Back - Returns to the previous step if you need to modify earlier settings.
- Continue - Moves to the next step of the wizard once the required fields are completed.
NOTE: If you click off the Wizard without Resetting or Canceling your inputs will be saved and you can return to complete the setup later.
Your Service is not created until you have been through all 6 steps and clicked the "Create Service" button.
Service Wizard Steps