Skip to content
Pilotoo
v0.1

0.1.0 Latest

ChangelogHow versions work

@pilotoo/voice-native

Introduction

Section titled “Introduction”

@pilotoo/voice-native runs speech-to-text and text-to-speech in Rust inside a Tauri 2 app. It contains the tauri-plugin-pilotoo-voice crate and its JavaScript side, which exposes a VoiceBackend for @pilotoo/voice. Voice activity detection and the speaker stay in the page. Use it in a desktop app that should transcribe and speak on the device.

Installation

Section titled “Installation”

@pilotoo/voice-native is at version 0.1.0 and is not published to a registry. It is also not part of the tarballs that pnpm release:check packs, so today it is used from the kit workspace:

Terminal window
pnpm install
pnpm --filter @pilotoo/voice-native build

Inside the workspace, depend on it with "@pilotoo/voice-native": "workspace:^". Add the tauri-plugin-pilotoo-voice crate to your Tauri app as well; the speech models are downloaded on demand.

Basic usage

Section titled “Basic usage”
import { createNativeVoiceBackend } from '@pilotoo/voice-native';


const backend = await createNativeVoiceBackend({ page });

page is the PageVoice your app supplies (see createWebPageVoice in @pilotoo/voice/web). The call asks the plugin for its models first, so it is async.

Entry points

Section titled “Entry points”
Import Provides
@pilotoo/voice-native The only entry point. createNativeVoiceBackend, nativeSynthesizer, nativeTranscriber, nativeModelStore, loadNativeModels, tauriIpc, composeModelStores.

Peer dependencies

Section titled “Peer dependencies”
  • @tauri-apps/api ^2.12.0

Learn more

Section titled “Learn more”