メニュー
AudioComponent on Specs — playback modes (LowPower/LowLatency), Mix-to-Snap, mic input profiles (Voice/Analysis, Echo Cancellation, Bystander Speech Rejection). Load for audio playback, ambient sound, button SFX, or mic / AsrModule use.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Lens Studio native packages: install from Asset Library, pack/unpack `.lsc`/`.lspkg`, bump versions, attach README/`.d.ts`, author setup scripts. APIs `NativePackageDescriptor`, `exportScript`/`exportPackage`, `unpack`, `instantiate`, `SetupScript`.
Migrate a Lens project from Spectacles (2024) to SPECS 27 APIs: WebView, Remote Media, Camera, Depth, Motion Controller, SyncKit, VoiceML→ASR, RemoteReferenceAsset→Supabase. Use when migrating or porting a Lens to SPECS 27, or about deprecated APIs.
Update outdated packages/libraries (SIK, UIKit, SSK, SyncKit) in a Lens Studio project via the Editor API. Use for 'update my packages', 'upgrade SIK', 'update dependencies', 'pull latest SIK', or any package update ask. Requires Lens Studio MCP.
Generate a 3D mesh, model, or asset for a Lens — build, create, or make props, characters, scenery. Four peer backends chosen per asset: SPECS text-to-3D (static default — best texture fidelity, async minutes-long jobs), FAST3D (faster static AI props), code-authored MeshBuilder via the mesh-builder-scripting skill (articulated, animated, parametric TypeScript content), and Blender voxel (blocky aesthetic; the only source of rigged GLBs). GLB backends return the file plus AABB metadata so the caller can size colliders and place the mesh in the world.
Generate musical phrases, chord progressions, beats, jingles, and short pieces via local algorithmic synthesis. Sample-free, license-clean, fully offline. Scale-degree composition, pattern-based arrangement, frequency-separated mixing rules. Outputs 44.1 kHz stereo WAV to Assets/GeneratedSFX/.
Generate non-pitched sound effects (UI, impacts, sweeps, foley, ambient textures, retro 8-bit) via local algorithmic synthesis. Outputs 44.1 kHz WAV to Assets/GeneratedSFX/. For musical phrases or chord progressions use /build-music instead.
| name | specs-audio |
| description | AudioComponent on Specs — playback modes (LowPower/LowLatency), Mix-to-Snap, mic input profiles (Voice/Analysis, Echo Cancellation, Bystander Speech Rejection). Load for audio playback, ambient sound, button SFX, or mic / AsrModule use. |
| user-invocable | false |
| paths | **/*.ts |
Specs applies several audio behaviors automatically that differ from the standard Lens Studio platform.
The Mix to Snap flag on AudioComponent (which routes audio into Snap recordings) is always on in Specs, regardless of the Inspector value. The flag is ignored on-device.
AudioComponent.playbackMode is only settable from script — there is no Inspector field for it. Specs default every component to Audio.PlaybackMode.LowPower.
LowPower (Specs default)Reduces device power use. Introduces playback latency. Use for ambient sound, background music, or anything where ~tens of ms of delay is acceptable.
LowLatencyMinimizes latency at the cost of higher power use. Use for immediate auditory feedback — button-press SFX, hit confirmations, anything tied to user input.
// Set in an OnStartEvent handler (not onAwake — property writes belong in OnStartEvent)
// ambient / background:
this.audio.playbackMode = Audio.PlaybackMode.LowPower
// button SFX / user-input:
this.audio.playbackMode = Audio.PlaybackMode.LowLatency
Full copy-paste class skeletons: references/audio-playback-examples.ts
Specs auto-applies microphone profiles based on which features the Lens uses. No profile is applied by default — a profile activates only when a feature that needs it is present.
| Profile | Filters applied | Triggered by |
|---|---|---|
| Analysis | Echo Cancellation (removes speaker output from mic signal) | Microphone Audio feeds (e.g. audio-templates/audio-analyzer) |
| Voice | Echo Cancellation + Bystander Speech Rejection (ignores speech not from the wearer) | LensStudio:AsrModule use in the project (see ls-clad:specs-asr) |
If both Microphone Audio and AsrModule are used in the same Lens, the Voice profile takes precedence over Analysis.
ls-clad:build-sfx — generate sound effects to load into AudioComponentls-clad:build-music — generate music tracksls-clad:specs-asr — speech-to-text on Specs (activates the Voice audio input profile)