@reson8-labs/agents-plugin-reson8

Reson8 STT plugin for LiveKit Agents for Node.js.

A single reson8.STT class that adapts to how LiveKit uses it:

• Streaming (stream(), used by voice agents) connects to the turn-aware Turns endpoint. Reson8 detects conversational turn boundaries server-side: it emits a preflight transcript (PREFLIGHT_TRANSCRIPT — an eager guess that the turn is over) that your agent can start responding to, then confirms it as a final transcript — or cancels it if the speaker keeps talking. Great for low-latency voice agents.
• Batch (recognize()) transcribes pre-recorded audio and returns the full transcript.

Languages

de en es fr fy it nl pl pt sv — see Languages.

Leave language unset to auto-detect each utterance, or pin a code for better quality. A comma-separated list restricts auto-detection to those languages.

new reson8.STT();                      // auto-detect
new reson8.STT({ language: 'nl' });    // Dutch
new reson8.STT({ language: 'nl,en' }); // auto-detect, restricted to these

Installation

npm install @reson8-labs/agents-plugin-reson8
# peer dependencies (provided by your agent app):
npm install @livekit/agents @livekit/rtc-node

Usage

import * as reson8 from '@reson8-labs/agents-plugin-reson8';

const stt = new reson8.STT({
  apiKey: 'your-api-key', // or set RESON8_API_KEY
  // language is auto-detected when omitted; pass a supported code to pin it
});

With a Voice Agent

import { voice } from '@livekit/agents';
import * as openai from '@livekit/agents-plugin-openai';
import * as reson8 from '@reson8-labs/agents-plugin-reson8';

const session = new voice.AgentSession({
  stt: new reson8.STT(), // streaming + server-side turn detection
  llm: new openai.LLM(),
  tts: new openai.TTS(),
});

Transcribing a file

const event = await new reson8.STT().recognize(audioBuffer);
console.log(event.alternatives[0].text);

Configuration

| Option | Env var | Default | |---|---|---| | apiKey | RESON8_API_KEY | required | | apiUrl | RESON8_API_URL | https://api.reson8.dev | | language | — | null (auto-detect; one or more supported codes, e.g. "nl" or "nl,en") | | sampleRate | — | 16000 | | encoding | — | "pcm_s16le" | | channels | — | 1 | | customModelId | — | null (custom model for recognition biasing) | | includeTimestamps | — | false | | includeWords | — | false | | includeConfidence | — | false (batch recognition) | | includeLanguage | — | false (report detected language while streaming) |

STT.updateOptions({ ... }) changes settings at runtime; active streaming sessions reconnect automatically to apply them.

Development

npm install
npm run typecheck
npm run build