SRT · MPEG-TS
- Video
- H.264
- Frame rate
- 1920 × 1080 · 30 fps
- Bitrate
- 4.5 Mbps
- Audio
- AAC · 48 kHz
Callaba runs as software alongside your hardware encoder. For distinct destination profiles, create separate Restream jobs; Web Player ABR is a separate playback workflow. Codec, acceleration, and container support depend on the selected deployment and media profile.
What is live video transcoding?Review the live video transcoding workflow in the API documentation
Compare pricingStart with the source profile, choose the delivery changes a real endpoint requires, then verify the resulting output.
A transcoding job should exist for a named delivery requirement, not as a generic copy of the source.
Change only the media fields the receiving system cannot accept from the contribution feed.
Yes, when the selected profile and deployment have the required capacity. Create separate Restream jobs for distinct delivery profiles; use Web Player when the viewer needs an ABR playback workflow.
Keep transport conversion distinct from media conversion, then test the exact protocol handoff the endpoint expects.
This table separates the requested media change from the evidence needed to accept it.
| Capability | Supported behavior | Acceptance check |
|---|---|---|
| Normalize one contribution feed for delivery | Accept a stable live input, then shape supported codec, resolution, bitrate, frame-rate, and audio settings for the actual player, platform, or downstream system that receives it. | Inspect the resulting codec and measured output properties, then decode the stream at its actual destination. |
| Create output profiles as independent jobs | Create a separate Restream job for every destination profile. That keeps target settings and runtime evidence clear without making the source encoder publish a separate feed for each viewer profile. | Start every required job and confirm that its bitrate, frame rate, and delivery state remain independent. |
| Convert protocols at the workflow boundary | Keep contribution and delivery choices separate. For supported routes, receive the protocol that fits the source and hand off the protocol expected by a platform, player, or partner endpoint. | Publish through the chosen input transport and verify that the receiver accepts the configured output transport. |
| Use your existing hardware encoder | Keep cameras and field encoders focused on producing a clean contribution feed. Callaba receives that feed as a software layer and prepares supported delivery outputs alongside the capture hardware. | Compare the clean encoder contribution with the processed output instead of changing capture settings prematurely. |
| Run the same product in cloud or self-hosted | Validate the workflow quickly in cloud, or install Callaba on infrastructure you control when data location, capacity planning, or private networking requires a self-hosted deployment. | Measure CPU or accelerator load, memory, and network headroom with the planned number of concurrent jobs. |
| Validate the exact media profile before production | Codec, container, protocol, acceleration, and output support depend on the selected workflow and infrastructure. Test the real source and destination profile before committing a production event or migration. | Record the tested source, output profile, destination, and resource use as the production acceptance baseline. |
The useful proof is a reproducible source-to-destination run with measured output and known infrastructure.
Leave capture hardware responsible for a stable contribution feed; apply downstream profile changes in Callaba.
Choose the environment after measuring codec load, concurrency, network path, and data-location requirements.
Run the exact codec, resolution, frame rate, audio, and destination combination before the live event.
You have seen what the product does. These three guides take you into the exact controls, show what to connect next, and give you a practical check before the workflow goes live.
Use cloud for a bounded proof with a real feed, or size self-hosted infrastructure around the measured workload.
Start with a cloud deployment when you want to validate a real source, output profile, and routing path before planning fixed infrastructure.
Deploy Callaba on AWSInstall the software on infrastructure you control and size compute or acceleration for the codecs, output profiles, and concurrent channels you plan to run.
Install Callaba self-hostedUse Callaba API modules when ingest, routing, stream state, and delivery outputs need to become part of your operator panel or application. Define and test the media workflow in the product first; automate it second.
Live video transcoding decodes an incoming feed and creates a supported output profile with different media settings, such as codec, resolution, bitrate, frame rate, or audio. It is useful when the source format does not match a destination or when one source must serve multiple playback profiles.
No. Callaba is software. A camera, OBS, vMix, mobile app, or hardware encoder sends a contribution feed into Callaba, and the software prepares supported routing and delivery outputs.
Depending on the selected workflow and available infrastructure, a profile can adjust supported codec, resolution, bitrate, frame rate, pixel format, GOP, and audio settings. Validate the exact input and output combination before production.
Supported routes can separate the contribution protocol from the delivery protocol. This lets the source use the transport that fits contribution while the destination receives the handoff it expects.
Use cloud for a fast proof of value and variable workloads. Choose self-hosted when you need direct control over infrastructure, private networking, data location, or capacity planning. Test your real media profiles in either model.
Start with the cloud deployment guide or the self-hosted installation guide. Connect the real source, define the required destination profile, validate quality and resource use, and only then automate it through the API if needed.
Deploy a bounded test, inspect the delivered media and resource use, then expand only the profile that passed.