Every release needs a face. The cover is the first thing a listener sees, often before they hear a single note. On AetherWave there are actually two different engines that turn a song into visual art, and they solve two different problems. One lets you direct the look. The other listens to the recording itself. Here is how each one works under the hood.
Every release needs a face. The cover is the first thing a listener sees, often before they hear a single note. On AetherWave there are actually two different engines that turn a song into visual art, and they solve two different problems. One lets you direct the look. The other listens to the recording itself. Here is how each one works under the hood.
1. The Album Art app: you direct, the model renders
The Album Art generator is built for speed and control. You upload your track, type in the title and artist, pick one of eight art styles (Photorealistic, Abstract, Digital Art, Vintage, Minimal, Surreal, Anime, or Street Art), and choose a square or portrait format.
Behind the scenes, we assemble a detailed prompt from those choices and hand it to a diffusion image model. Your style pick is expanded into a rich visual instruction, so "Street Art" becomes urban energy with bold tags and dripping paint, while "Minimal" pulls in the opposite direction. The model generates an original 1024 by 1024 cover in seconds, and you can regenerate until one clicks.
The audio file is not analyzed to make the still image. Instead, it comes back in the final step: you can animate your cover into a short looping video and we merge it with your track, giving you a moving cover for a Spotify Canvas, a Reel, or a TikTok.
This approach is deliberate. When you already have a vision for your release, you want to steer the art, not hand that decision to an algorithm. Style in, cover out.
2. Soul Forge: the audio does the talking
Soul Forge takes the opposite path. Here, the recording itself is the input, and the visuals are a consequence of what the music actually sounds like.
When you upload a track, we run it through Librosa, an audio analysis library, in a Python service. Librosa measures the real characteristics of the audio: tempo in BPM, musical key, overall energy, brightness, textural density, and the instruments it can detect in the mix. A slow, dark, sparse ballad and a fast, bright, dense club track produce very different numbers.
That analysis becomes the seed for a full band identity. The tempo, key, energy, and detected instruments feed the generation of a band name, its members, its backstory, and its collectible card art, so the visual identity reflects how the song truly sounds rather than a label you typed in. A high-energy, high-brightness track forges a very different band than a soft, low-brightness one, because the underlying audio data is different.
Audio analysis is required for Soul Forge by design. We would rather fail cleanly than invent an identity from guesswork, so if the analysis cannot run, the forge stops.
Two problems, one goal
Both tools exist to solve the same creative bottleneck, turning sound into something you can see, but they start from opposite ends. The Album Art app starts from your intent and renders it. Soul Forge starts from the audio and reveals an identity hidden in it. One is a design tool. The other is closer to a translator.
Try them both with a track of your own. Make a cover in seconds at www.aetherwavestudio.com/create-album-art, or forge a band identity from your sound at www.aetherwavestudio.com/static/forge-your-identity.html