MAI Realtime 1: what Microsoft's voice model does well, and where it stalls

Riellvriany Indriawan
Written by

Riellvriany Indriawan

Katelin Teen
Reviewed by

Katelin Teen

Last edited November 18, 2026

Expert Verified
Hand-drawn illustration of two people on a phone call and a laptop, with sound waves and the Microsoft logo on a blue shape
TL;DR

MAI Realtime 1 is Microsoft's first speech-to-speech model, and it is strongest at talking and weakest at finishing. It scored 98.9% on Big Bench Audio, the best on Microsoft's launch chart, but completed 57.5% of τ-Voice customer-service tasks, sixth of 13 models. At $1.35 an hour it is cheap, but it is a limited preview, so pilot it with a human handoff before it answers a real phone line.

That handoff is the part eesel covers: an AI helpdesk teammate that works the ticket queue behind the call. With the eesel CLI, a coding agent can load your call scripts and policies into it, then test answers before go-live.

What is MAI Realtime 1?

MAI Realtime 1 is a speech-to-speech model from Microsoft AI, the group behind MAI Transcribe 2 and the MAI Voice models. It takes in audio and answers in audio in one step, so there is no transcript-to-text-model-to-voice relay in the middle. Microsoft launched it at Ignite on November 17, 2026 and calls it its first speech-to-speech model.

MAI Labs page with the MAI-Realtime-1 model selected and a Start a conversation button, as taken from MAI Labs
MAI Labs page with the MAI-Realtime-1 model selected and a Start a conversation button, as taken from MAI Labs

I'm on the customer support side at eesel, so the first thing I ask about any voice model is whether it can run a call to the end, not whether it sounds nice. Here is the short fact sheet before I get to that.

QuestionAnswerSource
What is it?Speech-to-speech model for voice agentsLaunch post
Price$0.01 per audio minute, plus reasoning at $1 / $5 per 1M input / output tokensLaunch post footnote
AccessLimited Foundry preview and MAI Labs; no billing until public previewLaunch post
APICompatible with the OpenAI Realtime APILaunch post
Named testersAgoda and Reservations.aiLaunch post
Coming toDynamics 365 Contact Center, and Dragon Copilot for simulated patient conversationsLaunch post
Not publishedLatency number, language list, voices, regionsLaunch post

That last row matters. The launch post gives benchmark charts and a price, but no latency figure, language list or voice list, so anything beyond the table is yours to measure in the preview.

How does MAI Realtime 1 work?

It runs two layers. A fast voice layer keeps the conversation moving, and when a request needs deeper thinking, Microsoft says the model hands the turn to "a more powerful reasoning engine" so the answer still arrives without a long pause. The voice layer is the $0.01 minute rate. The reasoning engine is the separately billed token line, which I broke down in my MAI Realtime 1 pricing post.

Flow showing a caller's speech going to a fast voice layer at $0.01 per minute, which answers easy turns at once and sends hard turns to a reasoning engine billed per token, with a dotted line to your tools and systems
Flow showing a caller's speech going to a fast voice layer at $0.01 per minute, which answers easy turns at once and sends hard turns to a reasoning engine billed per token, with a dotted line to your tools and systems

Microsoft lists five behaviors it built for, and they read like a checklist of what makes phone calls awkward:

  • Natural turn-taking. It handles interruptions, quick acknowledgments like "mm-hm," and knows when to wait for someone to finish.
  • Noise handling. It recognizes requests with background noise or other people talking.
  • Real-world action. It connects to enterprise systems and tools, so a voice agent can do more than answer questions.
  • Personas. Tone, style and behavior change through prompting.
  • Adaptive reasoning. The fast-then-deep split above.

Microsoft's model page shows one worked example: a flight booking from Paris to Rome that runs 1 minute 21 seconds. The caller asks for Thursday, nonstop, window seat, under $200, then switches to Friday mid-call and says "same preferences and budget." The agent keeps all three constraints and quotes a $164 fare. That is a good demo of context carrying across a change of plan. It is a scripted demo, so treat it as the best case.

For developers, the part that lowers switching cost is compatibility with the OpenAI Realtime API. If your app already speaks that format, testing the model is mostly an endpoint change, though I would re-test tool calls on your own flows, since that is where voice agents usually break.

What is MAI Realtime 1 good at?

Spoken reasoning, by a clear margin. On Big Bench Audio, which asks reasoning questions out loud, MAI Realtime 1 scored 98.9% at $1.35 an hour on a fixed 40-question subset. Nothing else on the chart beat it, and the next best, Eleven Realtime v1 Max, scored 98.8% at $4.80.

Price is the second strength. Only Gemini 3.8 Live at $0.84 an hour is cheaper on Microsoft's chart. GPT-Realtime-2.1 High costs $10.75 an hour, nearly 8 times as much, and scored 95.8%.

Two caveats come with both numbers. The MAI figures are from Microsoft's internal evaluation, while the rivals' scores and costs come from Artificial Analysis. Microsoft also notes that results vary by evaluation setup and workload. Good signs, not a verdict. An independent run will tell us more.

Where does MAI Realtime 1 fall short?

On finishing the job. The second benchmark in the launch post, τ-Voice, simulates a customer calling a service desk to get something done, like changing a booking or processing a return. MAI Realtime 1 completed 57.5%, sixth of 13 models. Ahead of it: Phonic 1 High (79.9%), GPT-Live-1 Astra (74.5%), Grok Voice 3.0 (71.6%), Gemini 3.8 Live Extended Thinking (68.6%) and GPT-Live-1 Sol (59.3%).

The more interesting finding is that a high spoken-reasoning score does not predict a high task score. I lined up every model on the chart that has both numbers.

ModelBig Bench Audioτ-VoiceGap
GPT-Live-1 Astra (medium)90.1%74.5%15.6 pts
Gemini 3.8 Live Extended Thinking97.7%68.6%29.1 pts
Eleven Realtime v1 Mini75.9%46.8%29.1 pts
GPT-Live-1 Sol (low)89.0%59.3%29.7 pts
MAI Realtime 198.9%57.5%41.4 pts
GPT-Realtime-2.1 Mini High74.7%29.4%45.3 pts
GPT-Realtime-2.1 High95.8%45.7%50.1 pts
Eleven Realtime v1 Max98.8%47.6%51.2 pts
Gemini 3.8 Live91.7%30.1%61.6 pts

Scores from Microsoft's launch post; gap is my subtraction. Competitor figures are Artificial Analysis; MAI Realtime 1 is Microsoft's internal evaluation. τ-Voice covers full-duplex models only, so Big Bench Audio rows without a τ-Voice score are left out.

Dumbbell chart of five voice models showing a spoken reasoning score and a task completion score, with MAI Realtime 1 at 98.9% and 57.5%, a 41-point gap
Dumbbell chart of five voice models showing a spoken reasoning score and a task completion score, with MAI Realtime 1 at 98.9% and 57.5%, a 41-point gap

Every model drops from the first benchmark to the second. MAI sits mid-table on the gap: the highest spoken-reasoning score on the chart, paired with a middling task score. GPT-Live-1 Astra has the smallest gap, 15.6 points. I would not read this as a flaw in MAI so much as a reminder that spoken reasoning and completing a tool-driven task are different skills, and the support job needs the second one.

What does 57.5% mean on a support phone line?

Plan for about four in ten simulated tasks not completing. I would not apply that figure to your own calls, because τ-Voice uses simulated callers and set tasks, and Microsoft's number is its own. But it is the only task-completion figure on the table, and it is enough to design around.

In my day-to-day on the support queue, an unfinished call is rarely the real problem. The problem is the unfinished call that reaches a person with no context, so the caller has to start over.

Here is the pattern I would build around any voice model in this range:

Flowchart: call comes in, voice agent takes it, asks whether it can finish the task; if yes it finishes and logs a ticket, if not sure it does a warm handoff to a person with transcript and ticket attached
Flowchart: call comes in, voice agent takes it, asks whether it can finish the task; if yes it finishes and logs a ticket, if not sure it does a warm handoff to a person with transcript and ticket attached

The decision that matters is the diamond. A buyer in an eesel sales call, a CX lead handling around 7,000 tickets a month, put it this way:

"I need an AI who is only handling the tickets that it's confident to handle and all the other ones, leave them alone."

A CX lead running about 7,000 tickets a month on Gorgias, from an eesel sales call, March 2026

That is the whole design brief for a voice line too. Phone support is the channel where a wrong or half-finished answer is worst, because the caller is waiting in real time, so the handoff needs a transcript and a ticket attached, not a cold transfer.

BEAUTEK, a German equipment supplier whose own phone line is staffed 8:30 to 16:30 on weekdays, runs eesel as a live chat on its site on top of Zendesk. In one month (8 August to 8 September 2026) eesel answered 77 website chats and created 20 Zendesk tickets from chat for a person to pick up, with the customer's details and the conversation attached. The same shape works behind a phone model: the AI takes what it can finish, and everything else arrives as a ticket someone can pick up.

Who should use MAI Realtime 1 today?

Right now, teams that can join the preview and already have a human fallback. It is not open for production use. Microsoft's route in is a limited-access Foundry preview, and the public demo in MAI Labs needs sign-in, an 18+ confirmation, runs sessions of up to 2 minutes, and says it is for personal use only, not for commercial, school or work use.

If you are...My take
Already on Microsoft Foundry or Dynamics 365 Contact CenterRegister for the preview now. The model is coming to Dynamics 365 Contact Center, and you will be first in line
Building on the OpenAI Realtime APITest it as a drop-in on one non-critical flow, and measure latency yourself
Running a small support phone lineWait for public preview. A price of $0.01 a minute is not the bottleneck; handoff design is
Doing booking or routing callsGood first test. The Paris-to-Rome booking demo is the closest published example
Needing refunds or account changes by voicePilot only with a human approving, given the 57.5% task score
Comparing voice models on cost per finished taskRead my pricing breakdown; on Microsoft's numbers it is the cheapest per completed task

I would not pick it today if you need a language list or voice list, because the launch post does not publish one. Ask Microsoft in the preview form before you commit.

The wider field is moving fast. GPT-Live-1, Grok Voice 3.0 and ElevenLabs all ship voice agents today, and my AI voice companies comparison covers packaged options. The MAI launch post reached Hacker News on launch day with 3 points and no comments, so there is no practitioner feedback to lean on yet. The first people who run it on real calls will set the story.

Try eesel behind your phone line

A voice model is one layer of a phone line. The other layer is what the agent knows and where the call lands when it can't finish. eesel's AI helpdesk teammate joins your helpdesk, learns from your past tickets and help center, and works the queue the phone calls spill into. It hands off to your team with the conversation attached, and you can test it on historical tickets before it goes live.

If you are weighing AI for phone support, start with the queue, then add voice once the answers hold up.

Frequently Asked Questions

What is MAI Realtime 1?
MAI Realtime 1 is Microsoft's first speech-to-speech model, announced at Ignite on November 17, 2026 and described in its launch post. It listens and answers in audio directly, without a separate transcription and text-to-speech chain. My best AI voice agents roundup shows where packaged voice agents sit next to raw models like this one.
Is MAI Realtime 1 available to use right now?
Only in a limited way. Developers can register for a limited-access preview in Microsoft Foundry, and anyone can try a two-minute demo in MAI Labs, which is for personal, non-commercial use. There is no open sign-up for production traffic yet, so plan a pilot, not a launch. MAI Realtime 1 pricing covers what billing looks like once the preview ends.
How accurate is MAI Realtime 1 on customer service tasks?
On τ-Voice, a benchmark of spoken customer-service tasks, Microsoft's own run scored 57.5% task completion. That puts it sixth of 13 models on the launch chart, behind Phonic 1 (79.9%) and GPT-Live-1 Astra (74.5%). It scored 98.9% on Big Bench Audio spoken reasoning, so it answers questions well but finishes multi-step tasks less often. My guide on whether AI can answer support phone calls covers how to plan for the gap.
Does MAI Realtime 1 work with the OpenAI Realtime API?
Microsoft says it is compatible with the OpenAI Realtime API, so existing voice apps should be able to test it with limited rework. If you built on the OpenAI Realtime API, an endpoint swap is the first thing to try. Check realtime tool calls behavior on your own flows before you trust it.
Who should use MAI Realtime 1 for customer support?
Teams already building on Microsoft Foundry or Dynamics 365 Contact Center, with a human handoff ready, are the natural early testers. If you run a Zendesk contact center or a small phone line, wait for public preview and a latency number you can measure. A phone receptionist job, such as booking and routing, is the easier first test.
How does MAI Realtime 1 compare to Grok Voice 3.0 and GPT-Live-1?
On Microsoft's chart, MAI Realtime 1 costs about $1.35 an hour against $4.80 for Grok Voice 3.0 and $5.83 for GPT-Live-1 Astra, but both score higher on task completion (71.6% and 74.5%). MAI's own figures are an internal evaluation, and the rival figures come from Artificial Analysis. See Grok Voice 3.0 pricing for the per-call math.

Share this article