Audio 2 items

Everything Audio

🧠 DeepMind 4d ago

Gemini 3.1 Flash TTS: the next generation of expressive AI speech

DeepMind's Gemini 3.1 Flash TTS introduces granular audio tags for precise control over expressive speech synthesis. Enables directing AI-generated voice with fine-grained attributes for natural, controllable audio generation.

Multimodal Audio Models

📝 Blog 3w ago

Latent Space: Mistral Voxtral TTS - Flow Matching and Audio AI

Mistral's Voxtral uses flow matching for text-to-speech, expanding beyond text into multimodal audio. Discusses enterprise deployment and open source philosophy for audio models. Represents shift in how TTS will be productized and what "open" means for audio.

Multimodal Audio Open Weights

Gemini 3.1 Flash TTS: the next generation of expressive AI speech ↗

Latent Space: Mistral Voxtral TTS - Flow Matching and Audio AI ↗

Gemini 3.1 Flash TTS: the next generation of expressive AI speech

Latent Space: Mistral Voxtral TTS - Flow Matching and Audio AI