Microsoft releases MAI-Voice-2 for Foundry

// 2h agoMODEL RELEASE

Microsoft releases MAI-Voice-2 for Foundry

Microsoft has launched MAI-Voice-2, an expressive text-to-speech model featuring zero-shot voice cloning across 15 languages, now available on Azure AI Foundry for $22 per million characters. The model offers a cost-effective alternative to OpenAI's Realtime API and is integrated across VS Code, Teams, and Dynamics 365.

// ANALYSIS

Microsoft is positioning MAI-Voice-2 as a cost-effective, high-fidelity alternative to OpenAI's Realtime API, leveraging its Azure AI Foundry ecosystem to make natural voice synthesis accessible to developers.

* The $22 per million character pricing drastically lowers the barrier to entry for building production-grade voice applications.

* Support for 15 languages and zero-shot voice cloning enables seamless localization of voice agents.

* Vertical integration across VS Code, Teams, and Dynamics 365 ensures instant distribution and usage across Microsoft's massive enterprise footprint.

// TAGS

ttsspeechvoice-cloningmicrosoftazure-ai-foundryvoice-agentsdevtool

DISCOVERED

2h ago

2026-06-05

PUBLISHED

7h ago

2026-06-05

RELEVANCE

8/ 10

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE51m ago

Anthropic adds 'Dreams' to Managed Agents API

Anthropic launched "Dreams", an asynchronous memory consolidation feature for the Claude Managed Agents API. The offline job prunes stale context and resolves contradictions, resulting in a cleaner and more performant long-term memory store.

MODEL1h ago

Gemma 4 12B Assistant accelerates decoding

The Gemma 4 12B IT Assistant checkpoint has caused developer confusion, as it is actually a speculative decoding drafter model rather than a standalone chatbot. When configured in tools like vLLM alongside the primary model, it accelerates inference via Multi-Token Prediction without degrading quality.

BENCHMARK1h ago

Google's Gemma 4 12B model faces criticism for slow local performance and poor multimodal generation on Apple Silicon.

A developer tested Google's Gemma 4 12B model on a MacBook Pro M5 Max with 128GB of unified memory, finding its performance disappointing for a model of its size. Achieving only 44 tokens per second, the developer noted that the model's generation quality was subpar—specifically criticizing a produced lava lamp image—and recommended developers stick with Qwen 3.6 27B or 35B instead.