Tuesday, September 23, 2025

Microsoft AI Introduces Two Groundbreaking In-House Models to Power Its AI Mission

Related stories

Aisles Launches DREAM: AI-Driven Virtual Reality Evolution

Aisles has unveiled DREAM (Dynamic Reality Experience and Memory),...

TechSee Unveils Visual Remote Assistance with AI (VRAi) on Salesforce

TechSee, a global leader in visual customer assistance, announced...

Rendever and Lenovo Collaborate to Bring Virtual Reality Experiences to Carolina Caring Seniors

Rendever, the Boston-based company pioneering the future of aging...

Ansys 2024 R1 Reimagines the User Experience while Expanding Multiphysics Superiority Boosted by AI

The latest release from Ansys, 2024 R1, introduces an elevated user...

eXeX and Neurosurgeon Dr. Robert Masson Achieve World First Using Apple Vision Pro

eXeX™, a leader in artificial intelligence and mixed reality...
spot_imgspot_img

Microsoft AI unveiled two purpose-built, internally developed AI models that underscore its commitment to empowering every person and organization on the planet through applied, trustworthy AI technologies.

The first model, MAI-Voice-1, is a high-efficiency, highly expressive speech generation model now powering the Copilot Daily and Podcasts features. It is also available for experimentation via Copilot Labs, providing vivid, natural-sounding audio output across both single- and multi-speaker scenarios. The company emphasized that MAI-Voice-1 is a lightning-fast speech generation model, with an ability to generate a full minute of audio in under a second on a single GPU, making it one of the most efficient speech systems available today.

Also Read: TransPerfect Acquires Unbabel for Language AI

The second model, MAI-1-preview, represents Microsoft’s first end-to-end trained foundation model and is currently undergoing public testing on the AI benchmarking platform LMArena. This mixture-of-experts model has been both pre-trained and post-trained using approximately 15,000 NVIDIA H100 GPUs. Microsoft stated that MAI-1-preview is an in-house mixture-of-experts model, pre-trained and post-trained on ~15,000 NVIDIA H100 GPUs and is designed to provide powerful capabilities to consumers seeking to benefit from models that specialize in following instructions and providing helpful responses to everyday queries.

Microsoft AI further announced plans to roll out MAI-1-preview across certain text-based use cases within Copilot over the coming weeks, in order to gather user feedback and continuously enhance model performance.

“With this announcement, we are taking the first steps toward making our vision a reality,” Microsoft AI said. “We have big ambitions for where we go next. Not only will we pursue further advances here, but we believe that orchestrating a range of specialized models serving different user intents and use cases will unlock immense value. There will be a lot more to come from this team on both fronts in the near future. We’re excited by the work ahead as we aim to deliver leading models and put them into the hands of people globally.”

Subscribe

- Never miss a story with notifications


    Latest stories

    spot_img