The symbol description standard
An AAC symbol is more than a picture. In AAC Standard every symbol has a stable identity and structured metadata, so applications can use the same library in a consistent way.
1. Identification
Every symbol has two identifiers:
- a numeric ID (e.g.
1842) — the primary key in the API; never reused, - a slug (e.g.
apple) — a readable, stable identifier used in URLs; it never changes, even when the displayed name or translations do.
2. Metadata
| Categories (tags) | thematic groups of symbols, e.g. food, emotions, activities |
|---|---|
| Age group | child (0–6), youth (7–12, “junior”), adult — the same concept has separate graphic variants matched to the user's age |
| Key color | a color coding the part of speech on AAC boards (Fitzgerald-key style) |
| Grammatical category | pronouns, verbs, nouns (places/things), adjectives, conjunctions… |
| Person variants | for symbols depicting people: skin tone (5 steps), hair color, a monochrome version, overlays (plural, negation) |
3. Languages
Symbol names are translated (currently 9 languages, including Polish, English, German, Ukrainian, French, Spanish, Italian, Czech and Slovak). Audio recordings (TTS) are generated per language as well.
4. Quality
Every generated image passes an automatic meaning-accuracy check (scoring); results below the threshold go to human moderation. For ambiguous words the standard provides a gloss — a short definition of the intended meaning that the image is judged against.
5. Versioning and roadmap
The library holds 21061 symbols today and is growing towards 15,000 concepts. A formal asset-versioning specification and a public changelog of the standard are in preparation — until then, stability is guaranteed by immutable identifiers and the rule that a published slug never disappears.