How to push audio from Node.js and use Symbl real-time API to subscribe to real-time data

Getting started#

This example runs on node server, so we will use symbl-node package.

Open .env file and add your APP_ID, APP_SECRET, SUMMARY_EMAIL. You can get APP_ID and APP_SECRET from https://platform.symbl.ai

  require('dotenv').config()
  const {sdk} = require('symbl-node')

For demo purposes, we're using mic to simply get audio from microphone and pass it on to websocket connection

We use mic module so make sure you check that all requirements for the package are met. For instance we need sox package or ALSA tools to be installed.

mic installation instructions: https://www.npmjs.com/package/mic#installation

on Mac it can be as simple as running brew install sox

  const mic = require('mic')
  const sampleRateHertz = 16000
  const micInstance = mic({
    rate: sampleRateHertz,
    channels: '1',
    debug: false,
    exitOnSilence: 6,
  })

Initialize SDK#

  (async () => {
    try {
      await sdk.init({
        appId: process.env.APP_ID,
        appSecret: process.env.APP_SECRET,
        basePath: 'https://api.symbl.ai',
      })
    } catch (e) {}
  })()

We will also need a unique id to associate to our Symbl request. We will create this Id using uuid package

const id = uuid();

Now we can start the connection using sdk.startRealtimeRequest. We need to provide several important options to our realtime request.

insightTypes - we need to provide which insights will be detected. Supported ones are action_item and question.
config object with meeting title, confidence treshold and the sampleRate
speaker object to define who is the speaker. To distinguish between different speakers we also need to provide userId with valid email, so after meeting will end, we will receive summary email
handlers - these handlers are used to detect speach, messages and insights.
- onSpeechDetected - This will return live transcription of the call
- onMessageResponse - When processed messages are available, this callback will be called
- onInsightResponse - When Symbl detects an insight, this callback will be called.

So our complete code will look like this:

const connection = await sdk.startRealtimeRequest({
  id,
  insightTypes: ['action_item', 'question'],
  config: {
    meetingTitle: 'My Test Meeting',
    confidenceThreshold: 0.7,
    timezoneOffset: 480, // Offset in minutes from UTC
    languageCode: 'en-US',
    sampleRateHertz,
  },
  speaker: {
    // Optional, if not specified, will simply not send an email in the end.
    userId: 'valid@email.com', // Update with valid email
    name: 'My name',
  },
  handlers: {
    onSpeechDetected: (data) => {
      console.log(JSON.stringify(data))
      // For live transcription
      if (data) {
        const {punctuated} = data
        console.log('Live: ', punctuated && punctuated.transcript)
      }
    },
    onMessageResponse: (data) => {
      // When a processed message is available
      console.log('onMessageResponse', JSON.stringify(data, null, 2))
    },
    onInsightResponse: (data) => {
      // When an insight is detected
      console.log('onInsightResponse', JSON.stringify(data, null, 2))
    },
  },
})