Audial's Resynth tab has a switch above the uploader: Vital or Serum 2. The input rules are the same, the scoring is the same, and the job is the same: build an editable patch that reproduces one sound. What differs is the synth the patch is for, and that changes what it costs, where you run it and how long it takes.
The short answer
- Pick Serum 2 if you own Serum 2 and that is where your sounds live. The preset opens in Serum 2 itself with nothing extra to install.
- Pick Vital if you do not own Serum 2, want the whole chain free and open source, or want to resynthesise from inside your DAW without visiting a website.
If you cannot decide, run the same clip through both and keep the better one. That is two runs.
Side by side
| Vital | Serum 2 | |
|---|---|---|
| What you get | a .vital preset |
a .SerumPreset |
| Opens in | Vital, or Audial Synth, Audial's free GPLv3 fork of Vital | Xfer Serum 2, version 2.1 or newer |
| The synth | Vital has a free version, and Audial Synth is free | a paid synth from Xfer; Audial does not sell or bundle it |
| Resynthesis from inside your DAW | yes, from Audial Synth's Resynth panel | no plugin: run it on the website, the API, the Python SDK or MCP, then load the file |
| Time per run | about 90 seconds plus 13 seconds per second of sample | about 2½ minutes, plus up to 2½ minutes if a machine has to start |
| Input | one isolated sound, up to 20 seconds | the same |
| Also returned | a render, keyboard renders at C3 and C5, a JSON report with a similarity score | the same |
| From code | sound2vital |
sound2serum |
| On Audial | the same plan: 2 free runs, then $10 a month or $100 a year | the same |
The same sound through both
The resynthesis page plays one target sound, a synth pluck made for the demo, next to the renders of the Vital preset and the Serum 2 preset Audial returned for it, unedited. Listen to the three together: it is the quickest way to hear what the two targets do with the same input.
On your own sounds, compare the scores. Both targets are scored the same way, against the same clip, so the number next to each download is a fair comparison between them. A score is not taste, though. If the lower-scoring patch is the one that sits better in your track, keep that one.
Where you run it
The biggest practical difference is the workflow.
Vital can stay in your DAW from start to finish. Install Audial Synth (VST3, AU and standalone on macOS; VST3 and standalone on Windows), open its Resynth panel, drop a one-shot on it, and the new patch is written to your user presets and loaded into the synth. Audial Synth plays like Vital without an account; only the Resynth step calls Audial and needs a run. You can also use the website or the API and open the .vital file in plain Vital.
Serum 2 is a round trip: upload on the website (or call the API, the SDK or the MCP server), download the .SerumPreset, and drag it onto the Serum 2 window or into the user presets folder that Serum 2's preset browser points at. There is no Audial plugin in this path, and no need for one: the file is in Serum 2's own format.
Speed
Vital runs scale with the length of the clip: about 90 seconds plus 13 seconds for every second of sample, so a 2-second pluck takes about two minutes. Serum 2 runs take about two and a half minutes, plus up to another two and a half when a machine has to start because nobody has used it for a while. The site header shows when a machine is starting, so you are not left guessing.
Either way, the advice on input is the same, and it matters more than the target: one isolated sound, 1 to 4 seconds, trimmed at the front and dry. A loop, a melody, a drum pattern or a full mix gives a poor preset on both.
Editing the result
Both targets give you a real preset, not a frozen sample, so the edits are the same ones you already know: slow the attack to turn a pluck into a pad, pull the filter envelope down to tame a harsh source, add unison for width, put an LFO on something that should move. Both synths show their modulation routing, so you can see what the engine set up and remove what you do not want.
Pick the synth whose editor you are faster in. A preset is a starting point, and the target you will actually open and tweak is the better one for you.
Open source, if it matters to you
Audial Synth is a separate GPLv3 fork of Vital with a Resynth panel added. It is not affiliated with or endorsed by the Vital project, and its source is public at github.com/AudialAI/sound2vital-gui. Serum 2 is a commercial product of Xfer's; Audial only writes files in its format.
From code
The two targets are two functions with the same shape:
import audial
audial.config.set_api_key("your_api_key")
audial.config.set_user_id("your_user_id")
vital = audial.sound2vital("pluck.wav")
serum = audial.sound2serum("pluck.wav")
print(vital["preset"], vital["scores"])
print(serum["preset"], serum["scores"])
Over REST they are POST /api/functions/run/sound2vital and POST /api/functions/run/sound2serum, each with the clip inline as original. Through the Audial MCP server they are the sound2vital and sound2serum tools.
Where to go next
Start from the Vital preset generator or the Serum 2 preset generator, or try one on a vocal: a Serum 2 preset from a vocal chop.