AMSTERDAM—swXtch.io, a global provider of software-defined networking solutions for media, has launched an AI-based broadcast engineer platform which it says is designed to remove the infrastructure and networking complexity behind live media workflows.
abbe removes the requirement for operators to configure networks, cloud infrastructure, compute resources, and AI software in media routing and interface workflows, according to the company.
abbe takes a conversational description of the desired outcome and deploys the necessary infrastructure, networking, and software required to make the workflow run and does so in the most economical way while meeting performance requirements.
abbe is a cloud-agnostic orchestration layer, able to deploy live media and AI workflows across any provider including Microsoft Azure, Oracle Cloud Infrastructure (OCI) and Amazon Web Services (AWS). Workloads are located based on user requirements including optimizations for cost and latency, without forcing the operator to design a different solution for each cloud environment.
The initial capabilities being shown at IBC focus on two areas where live media teams face significant operational complexity: managing live media transport and deploying AI into production workflows.
For live media networking, abbe provides SRT management and control, enabling operators to ingest, monitor and distribute streams globally without manually configuring the underlying infrastructure. abbe is designed to automate decisions around where and how live media moves, helping operators maintain reliable, high-quality feeds while reducing the manual intervention traditionally required to respond to changing network conditions.
For AI workflows abbe will demonstrate a live audio processing workflow using NVIDIA models and other select partners. An operator can describe the desired outcome and abbe stands up the cloud infrastructure, networking, security, and software required for the audio processing use case.
The future of abbe is much broader than the initial iteration being shown at IBC, swXtch.io said.
abbe treats AI models as interchangeable components within a live workflow. Operators can select and test different models, chain models together, compare performance, and swap from one model to another easily and quickly. abbe separates what an operator wants to accomplish from the increasingly complex infrastructure, networking, and knowledge required to accomplish it.
NVIDIA models provide the foundation for abbe's initial real-time AI capabilities and swXtch.io is already expanding abbe's support across additional NVIDIA AI models and capabilities over time.
“The AI models are becoming incredibly capable but putting them into a live production still requires expertise across networking, cloud infrastructure, and AI software,” said Brent Yates, CEO of. “abbe is designed to remove that complexity. You describe what you want to accomplish, and abbe stands up everything needed to make it work. And because workflows are not locked to one model, you can easily optimize it as the technology changes. abbe simplifies AI for live media.”
