Speechmatics - API for Voice AI Agents
Mini toolsWavFlow
WavFlow is a Meta AI research model for multimodal audio generation in raw waveform space. It generates synchronized, high-fidelity audio from video and text inputs without relying on latent audio compression. Its method uses waveform patchifying and amplitude lifting to make flow matching stable directly on raw audio through direct x-prediction. The project reports evaluation on VGGSound for video-to-audio and AudioCaps for text-to-audio, positioning WavFlow as an end-to-end waveform-generation alternative to latent-based audio generation systems.
New
Multimodal
Gen 3
Overview
WavFlow is Meta AI’s open-source audio-generation model for creating synchronized high-fidelity audio from video and text directly in raw waveform space.
About Meta Platforms
We're connecting people to what they care about, powering new, meaningful experiences, and advancing the state-of-the-art through open research and accessible tooling.
Industry:
Technology, Information and Media
Company Size:
78865
Location:
Menlo Park, California, US
Website:
ai.meta.com
Tools using WavFlow
No tools found for this model yet.
Last updated: May 22, 2026
