Microsoft Build 2026: MAI models in Microsoft Foundry across text, image, voice, and speech

Learn Microsoft AI covers Microsoft Build 2026 announcements for MAI models in Azure AI Foundry, spanning text, image, voice, and speech modalities.

Overview

At Microsoft Build 2026, Microsoft announced updates to its first-party model lineup in Microsoft Foundry, expanding coverage across four modalities:

• Text
• Image
• Voice
• Speech

The video highlights the following models and their intended use cases for developers building AI applications:

• MAI-Thinking-1: positioned for advanced reasoning and coding scenarios.
• MAI-Image-2.5: positioned for image generation and image editing.
• MAI-Voice-2: positioned for multilingual voice cloning and text-to-speech.
• MAI-Transcribe-1.5: positioned for speech-to-text, with support for 43 languages.

Links

• Channel membership: https://www.youtube.com/channel/UCQf_yRJpsfyEiWWpt1MZ6vA/join
• LinkedIn: https://www.linkedin.com/in/rvinothrajendran/
• GitHub: https://github.com/rvinothrajendran
• Buy me a Coffee: https://buymeacoffee.com/vinothrajendran