Iron's Spells: Incantation

# Iron's Spells: Incantation **Say a spell's name. Cast it.** Incantation captures your microphone directly, runs the audio through an offline speech recognizer (Vosk), and casts whichever Iron's Spells 'n Spellbooks spell you just said. No voice training, and the speech model installs itself on first launch. --- ## How it works Incantation opens your microphone itself — no voice-chat mod involved — and feeds the audio straight to a Vosk recognizer whose grammar is built from the spell names your installed mods actually register. Anything that isn't a spell name resolves to "unknown" rather than snapping to the nearest spell, so background noise doesn't fire anything. Casting is then gated on what you actually have equipped: a spellbook in your Curios slot, or an imbued weapon you're holding. Saying the name of a spell you don't have does nothing. Servers that would rather let you cast from a spellbook held in hand can set `castMode = ANY_SPELLBOOK`. Recognition runs entirely on your machine. No accounts, no telemetry, and your microphone audio is never transmitted, never recorded, never written to disk. The one time Incantation touches the network is the speech-model download on first launch; set `autoDownloadModel = false` and install the model yourself if you would rather it didn't. ## When it listens **By default the microphone is only open while you have a spellbook, staff or imbued weapon — in either hand, or in your Curios slot.** Stow it and the mic closes. No keybind involved. That means conversation with your spellbook stowed can never cast anything, which matters because a lot of people run a voice-chat mod on the same microphone. **But if you keep a spellbook in your Curios slot permanently, this is close to `ALWAYS_ON` in practice** — so if you talk to friends mid-fight, use `HOLD_KEY_AND_ITEM`. Carrying a spellbook isn't on its own evidence you meant to cast; that mode requires a held key too, so speech never reaches the recognizer unless you asked. Set `gatingMode` to taste: | Mode | Microphone is open | |---|---| | `HOLD_ITEM` | while a spell focus is in your hands **or Curios slot** *(default)* | | `HOLD_KEY_AND_ITEM` | while holding the cast key **and** a spell focus | | `HOLD_KEY` | while holding the cast key | | `ALWAYS_ON` | whenever you are in a world — fully hands-free | These gate **capture**, not recognition: in every mode but `ALWAYS_ON` the recognizer never receives the audio, and the device is closed rather than merely ignored. The mic is also released whenever the game is unfocused or paused. The HUD's mic chip tells you which state you're in at a glance — the dot is dim when the mic is closed, lit when it's open, and pulses while it's actually hearing you, with a live level meter beside it. ## Features - **Speak any indexed spell** — works with Iron's Spells, all its addons, and any mod that registers spells through Iron's API. - **Equipped-only casting** *(default ON)* — a spell only fires if you actually have it on you. Anything else is rejected before it leaves your client. - **Imbued weapons supported** — hold an imbued sword or staff, say the spell it carries, it casts via the same mechanism as right-clicking. - **HUD chip** — corner-anchored chips showing a live mic indicator and level meter, plus last cast, recent cast history, queued spells, miss toast, the last phrase heard, and "did you mean…?" alias suggestions. - **Vanilla advancements** — voice-cast milestones (1, 10, 50, 200, 1000) and combo casts surface as standard advancement toasts. - **Spell List screen** — browse every spell the registry exposed with the exact phrase Vosk listens for, plus per-spell cast counts and school filtering. - **Voice Codex** — personal stats (top-cast spell, daily streak, median latency). - **Test Arena** — safe practice mode that records what *would* have cast without firing anything. Useful for tuning aliases or learning a tricky spell's pronunciation. - **Welcome wizard** on first launch — walks you through mic + model + first cast. - **Aliases & custom phrases** — bind words the model can pronounce to any spell id. - **Play in your own language** — Spanish, Russian and French speech models are one config line away (`modelId`), and `config/voicespells/phrasebook.json` is generated with every installed spell and an editable `override` field, so you can translate the whole spell list in one file instead of adding aliases one at a time. Your overrides survive updates, and newly installed spells are appended automatically. **The model is only half of it.** The recogniser only listens for words its model knows, so the spell phrases have to be translated too — a Russian model against English spell names hears nothing at all. A Spanish phrasebook contributed by the community, and a Russian one part-filled from Iron's Spells' own translation, are both in the repository's `contrib/` folder. **German does not work yet**, and not for want