API ReferenceLocal APIs

useLocalTTS

React hook — on-device text-to-speech (Kokoro) over WebGPU.

Auto-generated from JSDoc in packages/react/src/local/useLocalTTS.ts. Do not edit — run npm run docs:api.

useLocalTTS

function

function useLocalTTS(opts: UseLocalTTSOptions): UseLocalTTS

UseLocalTTS

interface

interface UseLocalTTS

`status: TTSStatus`

`progress: number`

`voices: string[]`

`error?: Error`

`audioUrl?: string`

Object URL of the most recent synthesized clip (revoked + replaced on each speak).

`load: () => Promise<SpeechModel>`

`speak: (text: string, voice?: string) => Promise<void>`

UseLocalTTSOptions

interface

interface UseLocalTTSOptions

`device?: unknown`

Default "webgpu". Falls back to "wasm" if you prefer CPU.

`dtype?: unknown`

`autoLoad?: boolean`

TTSStatus

type alias

type TTSStatus = idle | loading | ready | speaking | error

useLocalModel

React hook — on-device LLM chat over WebGPU.

useVAD

React hook — real-time mic voice-activity detection (Silero).

On this page

useLocalTTS UseLocalTTS status: TTSStatusprogress: numbervoices: string[]error?: ErroraudioUrl?: stringload: () => Promise<SpeechModel>speak: (text: string, voice?: string) => Promise<void>UseLocalTTSOptions device?: unknowndtype?: unknownautoLoad?: booleanTTSStatus

Valora is local-first

No API key, no server — everything in this doc runs on-device.