Link: Microsoft AI launches its first in-house models

Microsoft's AI division has launched two new AI models, MAI-Voice-1 AI and MAI-1-preview. The MAI-Voice-1 can create a minute of audio in under a second using a single GPU.

This speech model is utilized in features such as Copilot Daily and enables the generation of podcast-style content. Users can experiment with it on Copilot Labs, customizing the voice and speaking style.

Meanwhile, the MAI-1-preview, trained on 15,000 Nvidia H100 GPUs, is designed to assist users by responding to everyday inquiries. It will be integrated into Microsoft's Copilot AI assistant for text-based tasks.

Microsoft AI chief Mustafa Suleyman emphasizes that their AI developments aim to enhance consumer experiences rather than focusing solely on enterprise applications. These models leverage extensive consumer data to fine-tune their functionalities.

The company is also testing MAI-1-preview on LMArena, a platform for AI benchmarking. Microsoft's broader goal is to develop a variety of specialized models to address diverse user needs effectively.

As part of this vision, Microsoft acknowledges the immense potential of their approach in transforming AI interactions and enhancing model capabilities for various applications. #

--

Yoooo, this is a quick note on a link that made me go, WTF? Find all past links here.