Microsoft has launched VibeVoice, a new open-source AI model capable of generating up to 90 minutes of multi-speaker audio ...

Meta Platforms Inc.’s artificial intelligence research team today said it has open-sourced a new project called Massively Multilingual Speech, which aims to overcome the challenges of creating ...

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

OpenAI's Realtime API is now generally available, featuring the new gpt-realtime model for more natural voice agents at a 20% ...

It notes that smaller organizations have a distinct disadvantage when trying to develop models for speech recognition, because the most comprehensive datasets available have always had high ...

LAION, the German nonprofit behind a number of popular AI data sets, wants to develop an open voice assistant called BUD-E.

OpenAI's new tool in pilot, Custom Voices, allows developers to create a clone of any voice with a 15-second sample.

Saint Louis University’s (SLU) Reinert Speech-Language-Hearing Clinic offers gender-affirming voice training as a way for clients to work to align their voice and communication to better represent ...

The University of Colorado and members of the Trans Voice Initiative will offer a two-day gender-affirming voice training led by gender-diverse speech-language pathologists (SLPs). The course is open ...