AI ASSISTANT

Reactive Editor

Introduction

The AI assistant (Figure 1) is built into the Reactive Editor. You describe the reactive element you want in plain language, and it generates a node graph for you, which loads straight into the editor ready to run, inspect and edit.

It is there to save time, not to do anything you could not do yourself. Every graph it produces is an ordinary node graph made of the same nodes documented in the node reference, and once it has loaded you own it completely: change it, extend it, take it apart to see how it works, or ignore the result and build your own. Nothing in Sound to Vision is locked behind it, and nothing it makes is off limits to edit.

Where it tends to earn its keep is at the start of something, roughing out a structure you then refine, or working out how a particular effect might be approached when you are not sure where to begin. Plenty of people never use it at all, which is a perfectly good way to work.

How it works

Open the AI panel using the AI button on the main menu, or the toggle on the right of the screen. Type a description of what you want into the text entry and press send. A few moments later the assistant returns a node graph as JSON output; click that output to load it into the editor, then press Run to see it.

Two modes are available at the bottom of the panel. Start fresh generates a new graph from scratch and is the default. Modify makes changes to the graph you just generated, which is the better option for refining a result over several passes rather than trying to describe everything perfectly the first time.

Sound to Vision does not store your generated graphs. Anything you want to keep should be saved from the menu, exported as a JSON file, or loaded into the main Sound to Vision interface. A graph that is replaced by a fresh generation without being saved cannot be recovered.

Credits

Each request you send to the assistant uses credits. This is because every generation runs against a commercial language model that charges per request, and those costs are passed on rather than absorbed into the price of the software. Credits are how that is handled.

A few things worth stating plainly:

• Credits apply only to the AI assistant. Everything else in Sound to Vision, in both the browser and Pro editions, works exactly the same whether you have credits or not.
• There is no subscription. You buy credits when you want them and they sit on your account until used.
• Credits do not expire.
• A failed request does not consume credits.

You need an account to hold credits, so signing in is required both to buy them and to use the assistant. If you do not have an account yet, you can create one here.

Credit packs

Credits are sold in packs. Larger packs work out slightly cheaper per credit, which reflects the reduced payment-processing overhead rather than any pressure to buy more than you need. The smallest pack is a perfectly sensible place to start if you only want to try the assistant out.

Prices exclude tax, which is calculated at checkout based on your location. Payments are handled by Stripe. Sound to Vision never sees or stores your card details.

Common questions