Microsoft has announced the general availability of gpt-realtime, its latest speech-to-speech (S2S) model, on Azure AI Foundry.

Roughly two weeks ago, Google Docs gained a key feature that should make absorbing swaths of information an easier task. The ...

"VibeVoice is a novel framework designed for generating expressive, long-form, multi-speaker conversational audio, such as ...

Microsoft’s VibeVoice is an open-source text-to-speech model for podcast-length, multi-speaker audio that captures the authentic conversational “vibe.” ...

VibeVoice is a new open-source AI tool that can generate a full 90 minute audio podcast recording with multiple speakers from ...

Google is testing a feature that creates podcast-style summaries of your news feed right on Chrome's New Tab page ...

OpenAI’s GPT-Realtime is reportedly the company’s most advanced voice model, designed for customer support and assistance.