Microsoft AI (MAI) released MAI Realtime 1, its first speech-to-speech model, during Ignite on November 17. It targets enterprise voice agents, with base speech-to-speech usage at $0.01 per minute and reasoning billed separately at $1 per million input tokens and $5 per million output tokens. Microsoft says it beats GPT Realtime 2.1 on quality, speed and price across public benchmarks such as Big Bench Audio and tau-Voice and on Microsoft product evaluations. The model is compatible with the OpenAI Realtime API for easy migration; it can be tried in MAI Labs, and limited-preview customers in Foundry are not billed until public preview.
Key Takeaways
- ✓Pricing: base speech-to-speech usage is $0.01 per minute; reasoning is billed separately at $1 per million input tokens and $5 per million output tokens
- ✓Benchmarks: Microsoft says it sits on the performance-price Pareto frontier for Big Bench Audio speech reasoning and tau-Voice agentic task completion, and beats GPT Realtime 2.1 on quality, speed and price; the blog gives no exact scores
- ✓Capabilities: handles interruptions and quick acknowledgments and knows when to wait; a fast system keeps the conversation flowing while a stronger reasoning engine handles harder tasks; it copes with background noise and connects to enterprise systems and tools
- ✓Compatible with the OpenAI Realtime API, so existing voice apps can be tested and migrated with little work
- ✓Availability: try it in MAI Labs or join the limited-access preview in Microsoft Foundry (no billing until public preview); Agoda and Reservations.ai are testing it, and it is coming to Dynamics 365 Contact Center and Dragon Copilot

Key Decision Metrics at a Glance
Turn your technical choice into a development budget
Compare 40 dev plans & simulate token costs vs $20/mo subscriptions
Project Links & Resources
Direct AccessIn-Depth Technical Analysis
Background
Enterprise voice agents for customer service, bookings and training need natural, interruptible conversation that gets things done, at a low cost per minute. OpenAI's Realtime models have led this market. Microsoft AI used Ignite to launch MAI Realtime 1, its own first speech-to-speech model.
How it works
Microsoft has not published architecture or parameter details. It describes a fast system that keeps the conversation going in real time and calls a stronger reasoning engine when a task needs deeper thinking, avoiding long pauses. It handles interruptions, recognizes requests through background noise and connects to enterprise systems through tools. Tone and persona can be set by prompting.
Numbers that matter
Microsoft says MAI Realtime 1 beats GPT Realtime 2.1 on quality, speed and price on the public Big Bench Audio and tau-Voice benchmarks and in Dynamics 365 Contact Center evaluations, and sits on the performance-price Pareto frontier. The blog does not list exact scores.
What it means for developers
Because it is compatible with the OpenAI Realtime API, existing apps can switch over for testing. Try it in MAI Labs or request the limited preview in Microsoft Foundry, which is free until public preview; list pricing is $0.01 per minute with reasoning billed separately.
Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.