---
title: "Microsoft launches three in-house AI models for speech, voice and images"
description: "MAI-Transcribe-1, MAI-Voice-1 and MAI-Image-2 are now on Microsoft Foundry, a step toward relying less on OpenAI."
url: https://kurddevelopers.com/en/news/microsoft-mai-transcribe-voice-image-models
language: en # English
type: news
published: 2026-04-02T16:48:00.000Z
updated: 2026-11-02T03:03:34.000Z
author: "Osamah Ismail"
author_url: https://kurddevelopers.com/en/author/1
section: "AI"
tags: ["Microsoft","MAI","AI","Foundry"]
sources: ["Microsoft AI","TechCrunch"]
image: https://kurddevelopers.com/uploads/2026/11/social02-db3dfa56.jpg
translations:
  ku: https://kurddevelopers.com/news/microsoft-mai-transcribe-voice-image-models
  ar: https://kurddevelopers.com/ar/news/microsoft-mai-transcribe-voice-image-models
publisher: Kurd Developers
publisher_url: https://kurddevelopers.com/
---

# Microsoft launches three in-house AI models for speech, voice and images

> MAI-Transcribe-1, MAI-Voice-1 and MAI-Image-2 are now on Microsoft Foundry, a step toward relying less on OpenAI.

![Image: Microsoft AI](https://kurddevelopers.com/uploads/2026/11/social02-db3dfa56.jpg)

## What's new
The **Microsoft AI** team led by Mustafa Suleyman released three new models on **Microsoft Foundry**:

| Model | Task | Price |
|---|---|---|
| MAI-Transcribe-1 | Speech to text, 25 languages | from $0.36 per hour |
| MAI-Voice-1 | Voice generation | from $22 per 1M characters |
| MAI-Image-2 | Image generation | from $5 per 1M text input tokens and $33 per 1M image output tokens |

## Key points
- MAI-Transcribe-1 is **2.5x faster** than Microsoft's previous Azure Fast offering.
- MAI-Voice-1 can generate **60 seconds** of audio in **one second**.
- MAI-Image-2 first appeared on MAI Playground on March 19.

## Why it matters
Microsoft has invested more than $13 billion in OpenAI, but it is now building its own models to compete across the board.

Prices are in US dollars; prices in Kurdistan may differ.

**Sources:** Microsoft AI, TechCrunch

---

Published by [Kurd Developers](https://kurddevelopers.com/) on 2026-04-02. Cite as: “Microsoft launches three in-house AI models for speech, voice and images”, Kurd Developers, https://kurddevelopers.com/en/news/microsoft-mai-transcribe-voice-image-models