Microsoft is today introducing container support for Azure Cognitive Services in preview. This will allow users of Azure Cognitive Services’ premade models to deploy AI via Docker containers for edge devices, in the cloud, or in on-premise servers.
About 1.2 million developers have used Azure Cognitive Services, whose premade AI models can be adapted to meet the needs of companies and developers without the time or resources to make their own models from scratch.
Container support may appeal to those who find drawing AI predictions from the cloud expensive, businesses with privacy or data requirements, or anyone who simply doesn’t want to put their data in the cloud.
“It’s [containers] become the default way to deploy anything, and so AI models wrap up really quite nicely into a Docker container. That’s how people want to consume it, and so that’s the way that you should author them,” Microsoft VP Eric Boyd told VentureBeat in an interview. “We want to unblock where people are being blocked by from adopting AI. This is one of the key assets we’ve got, so that’s one of the things we’re doing.”
The release comes less than a week after Google Cloud released Kubeflow Pipelines, a machine learning workflow for Kubernetes containers to give companies and developers more options for deploying AI.
Facial recognition, vision, and speech-to-text services will first be available for containers, with more to come in the future, Boyd said. The three were chosen as the first of Microsoft’s more than 30 AI services for “a mix of technical and business reasons.”
“Some of the models are easier to … put into containers, and the other side is just really thinking through — What are people going to want and how are they going to use this and how is this going to evolve? It’s in preview, and we expect to learn a lot,” Boyd said.
Microsoft announced a series of AI-related stories today, including the acquisition of XOXCO, the creator of conversational AI bot design platform Botkit, and an automated AI model detection feature for Power BI software users.