/**
 * Audio-driven mouth animation addon for xrblocks. The primary export,
 * `LipsyncMouth`, is a small `xb.Script` that pulls audio from a
 * `MediaStream`, runs an FFT + formant-based viseme mapper on it every
 * frame, and writes the viseme weights to a `setVisemes`-compatible
 * target — typically the `face` on a netblocks `RemoteUserAvatar`, or
 * a standalone `xb.StylizedFace` for non-multiplayer use.
 *
 * The face primitive itself (`StylizedFace`) and the viseme contract
 * type (`VisemeWeights`) live in xrblocks core so neither addon has
 * to depend on the other.
 *
 * @see {@link LipsyncMouth}
 */
export {LipsyncMouth} from './LipsyncMouth';
export type {LipsyncMouthOptions, VisemeTarget} from './LipsyncMouth';
export {FormantVisemeMapper} from './FormantVisemeMapper';
export type {
  AudioFeatures,
  FormantVisemeMapperOptions,
} from './FormantVisemeMapper';
export {MfccExtractor, NUM_MFCC} from './MfccExtractor';
export type {MfccExtractorOptions} from './MfccExtractor';
export {
  ARKIT_BLENDSHAPE_NAMES,
  blendshapesToVisemes,
  ZERO_VISEME,
} from './BlendshapeReducer';
export type {VisemeWeights} from './BlendshapeReducer';
export {computeAudioFeatures} from './computeAudioFeatures';
export type {AudioFeatureInputs} from './computeAudioFeatures';