Sound is now on the canvas

The canvas has always been a place for thinking out loud.

It’s that early, exploratory part where you're trying things, following instincts, and seeing what a direction actually feels like before you've committed all the way. For a long time, that process was silent.

Sound is foundational to how creative work comes together, whether it’s a reference track playing while you concept, a voice that makes a character click, or a texture that tells you a direction is right before you can say why. Creatives have used audio to find their way for as long as there's been creative work. It just happened in another app, in another window, cut off from everything else on the canvas.

Audio Node changes that. Sound is now a new primitive on the canvas, and a node, it’s now something you can generate, connect, and build with, the same as any piece of text, image or video.

What that sounds like

The most requested version: you're making a video in FLORA and you need a voiceover. Write the script, pick a voice, generate it, play it back on the canvas, export it. No need to open a second tool, drag files between apps, or leave the canvas mid-exploration.

A more mind-opening application of Audio Node: what happens when audio becomes a part of how you explore a direction? Let's say you're developing a campaign in FLORA, and building toward something you can't fully describe yet. Now you can generate sound alongside text, image, and video. A voice that matches the character of a brand you're still figuring out, or foley for a video you haven't fully decided on.

But there's something else here that's harder to name. Sound has always been part of how products feel. More viscerally, it's the zip of a puffer jacket, the click of a well-made clasp, that very specific silence of a luxury car door closing. Those sounds are brand. They've just never been something you could explore on a creative canvas before. If you're building a campaign for a product, you can now ask: what does this idea sound like? What should the sonic identity of this brand direction be? That's a genuinely new question to be able to ask inside a tool like this.

What's included today

Text-to-speech via ElevenLabs, sound effects, lipsync for UGC workflows, audio transcription, and foley for video are all live now. If you need a consistent voice — yours, a client's, a character — you can train a custom voice directly in FLORA: upload files or record in-app. Plus, as your multi-faceted creative agent, FAUNA can build audio workflows the same way it builds anything else.

Bring your ideas to life in FLORA, now in even more ways.