Mistral released a new audio AI model called Voxtral. The French AI company stated that the model is geared towards ...

New research shows models can be directly edited to hide selected voices, even when users specifically ask for them.

French startup Mistral has jumped into the audio race with Voxtral, its first open model, aiming to challenge the dominance ...

‘Rumble, rumble, rumble, WHOOOOOOSH, *fiery launch noises*.’ ...

Both new models, Voxtral Small and Mini, include 32k token context lengths and built-in Q&A functions. They're both natively ...

First, Adobe Firefly allowed creators to generate AI video; now, the AI will help video editors generate audio too ...

I have created hundreds of hours of AI video content over the past couple of years including anime, short films and music videos. These are the best models I’ve tried.