Upload one isolated sound of up to 20 seconds to Audial's Resynth tab with Vital as the target (or drop it on the free Audial Synth plugin), and you get back a .vital preset whose oscillators, filters and envelopes are set to reproduce it, plus a render and a similarity score.
$10 a month or $100 a year. Unlimited runs, cancel any time. 2 free runs to try it first.
Audial's Vital preset generator gives you the patch behind a sound. Upload a single note or hit, keep Vital as the target on the Resynth tab, and Audial's hosted engine searches for the oscillators, filters, envelopes and effects that reproduce it, then hands you a .vital file. Because it is a preset and not a sample, nothing is baked in: slow the attack, open the filter, add an LFO, play it across the keyboard.
Open the preset in Vital, or in Audial Synth, Audial's free GPLv3 fork of Vital with a Resynth panel added. In Audial Synth the whole trip happens inside your DAW: drop a one-shot on the Resynth panel and the new patch is written to your user presets and loaded into the synth. Builds are on the downloads page for macOS (VST3, AU and standalone) and Windows (VST3 and standalone).
Every run is scored. With the preset you get a render of the patch, two keyboard renders (C3 and C5) and a JSON report with a similarity score between the render and your source. A job takes about 90 seconds plus 13 seconds per second of sample, longer if a machine has to start. From code it is /functions/run/sound2vital, audial.sound2vital in Python, or the sound2vital MCP tool.
What you get
- Any single sound up to 20 seconds becomes an editable .vital preset
- Opens in Vital, or in Audial Synth, a free GPLv3 fork of Vital
- Audial Synth runs the same resynthesis from a panel inside your DAW
- Returns the preset, a render, keyboard renders at C3 and C5 and a similarity score
- About 90 seconds plus 13 seconds per second of sample
- The same engine also builds Serum 2 presets
Hear it
The target is a synth pluck made for this demo. The render is the .vital preset Audial returned, unedited.
Audial vs doing it by hand
| Task | Audial | By hand |
|---|---|---|
| Getting close to the sound | Audial: The engine searches Vital's parameters and scores the match against your clip | By hand: Sound design by ear: pick a wavetable, shape the filter and envelopes, compare, repeat |
| Time | Audial: About 90 seconds plus 13 seconds per second of sample | By hand: Minutes for a simple pluck, much longer for a complex sound |
| What you end up with | Audial: A real .vital preset: every parameter editable | By hand: The same, and you learn the synth on the way |
| Sounds the synth cannot make | Audial: Gets as close as Vital can, and the score says how close | By hand: The same limit: you cannot patch what the synth cannot produce |
| Cost | Audial: 2 free runs, then $10 a month with unlimited runs; Audial Synth is free | By hand: Free apart from your time |
Use it from code
curl -X POST "https://api.audialmusic.ai/api/functions/run/sound2vital" \
-H "X-API-Key: your_api_key" -H "X-User-ID: your_user_id" \
-F "userId=your_user_id" \
-F "original=@/path/to/pluck.wav"import audial
audial.config.set_api_key("your_api_key")
audial.config.set_user_id("your_user_id")
result = audial.sound2vital("pluck.wav")
print(result["preset"]) # the .vital preset built from the clip
print(result["scores"]) # how close the render came to the clip# In Claude, Cursor or any MCP client with audial-mcp installed:
"Turn ~/Samples/pluck.wav into a Vital preset and tell me how close it scored."Frequently asked questions
Upload it on the Resynth tab with Vital as the target (the default) and run it. You get back a .vital preset: drop it into Vital's preset browser, or open it from Vital's Load menu. Or skip the browser: drop the sample on the Resynth panel in Audial Synth and the preset loads straight into the synth.
Audial Synth is a free download and plays like Vital without an account. Its Resynth panel calls Audial's hosted engine, so resynthesis needs an Audial account: 2 free runs, then a subscription at $10 a month or $100 a year.
One isolated sound: a held synth note, a pluck, a bell, a bowed note, a pad stab, a vocal ahh. 1 to 4 seconds is ideal and 20 seconds is the limit. Trim the silence off the front and keep it dry, because reverb in the clip gets modelled into the patch. A loop, a melody, a drum pattern or a full mix gives a poor preset, so cut one note out first.
Every run is scored against your clip, and the score is shown next to the download, so you can judge a patch before you open it. Clean single sounds score highest. A sound Vital cannot make will only ever get as close as Vital can.
It is a separate GPLv3 fork of Vital with a Resynth panel added, and it is not affiliated with or endorsed by the Vital project. A .vital preset from Audial opens in either one.
Yes. Post the clip to /functions/run/sound2vital, call audial.sound2vital in Python, or use the sound2vital tool through the Audial MCP server. Each returns the preset, the renders and the score.