Using Firebase Functions as a Conferencing Authentication Service
Fabien Lavocat

This blog post will show you how to use the Google Firebase Functions to build an authentication service with Node.JS (in TypeScript) for calling from each of the Android, iOS, and JavaScript Interactivity API SDKs. At, we consistently think about security and how to help our customers build applications and services that follow security best practices. As an example, we recently introduced the Enhanced Conference Access Control capability that allows you to set permissions for your users joining conferences.


In this blog post I want to focus on the initialization of the SDK in your applications. In the past, when you wanted to create an application that requires no backend to work, it was quite difficult to avoid hardcoding the Consumer Key and Consumer Secret. There are a few problems there. If your key and secret are hardcoded into your applications, you can consider that anyone (with minimal knowledge) can capture them, by either decompiling your application or by listening to the traffic going through the authentication endpoint. Now, lets say you need to use a new key for your applications, you will need to build a new version of your application and have all of your users upgrade at the same time if you want them to be able to communicate with each other (you cannot join a conference created with another consumer key). This is not practical for many users.

To avoid this, we recommend that our customers use a backend with a protected endpoint that will generate an access token that can be used by your applications to initialize the SDK. But creating and maintaining a server for only this endpoint can become very expensive and time consuming. You need to make sure it scales with the load of your application, you need failover, and you need to update the server with the latest patch updates. This is 2021 and all of the major cloud providers are now offering a serverless service; Azure FunctionsAWS Lambda and Google Cloud Functions. The advantage of running a cloud function for this kind of workload is that you do not have to maintain a server; you do not have to worry about scaling and the cost is minimal.

We are going to deploy a Firebase Function that we will use to generate an access token to initialize the SDK. Firebase Functions are built on top of the Google Cloud function. As of writing this post, Google Cloud does not charge for the first two million executions of your function (see the pricing page for more details). Note that the Google Cloud Functions are running from containers and there is a cost associated with hosting a container registry.

Firebase Function

Install the Firebase Tools on your machine using the following npm command (Node.JS):

npm install -g firebase-tools

Log the CLI into Firebase:

firebase login

Create a folder where you want to have your code and initialize your project with the following command:

firebase init

You will be asked what you want to set up in this folder, select Functions: Configure and deploy Cloud Functions. If you already have a Firebase project, select Use an existing project. Select the language you want to use to write the Cloud Function. I will provide you the code in TypeScript. Use ESLint to have a clean TypeScript code, then install the dependencies with NPM.

Open the functions/src/index.ts file. This will be the code of your function. Replace the content with the following code. Make sure to replace your CONSUMER_KEY and CONSUMER_SECRET. You can go one step further if you want to protect these by using the Secret Manager from Google Cloud.

import * as functions from "firebase-functions";
import * as http from "http";
import * as https from "https";
// Enter your Consumer Key and Secret from the dashboard
 * Sends a POST request
 * @param {string} hostname
 * @param {string} path
 * @param {OutgoingHttpHeaders} headers
 * @param {string} body
 * @return {Promise<any>} A JSON payload object through a Promise.
const postAsync = (hostname: string, path: string, headers: http.OutgoingHttpHeaders, body: string): Promise<any> => {
  return new Promise(function(resolve, reject) {
    const options: https.RequestOptions = {
      hostname: hostname,
      port: 443,
      path: path,
      method: "POST",
      headers: headers,
    const req = https.request(options, (res) => {
      functions.logger.debug(`[POST] ${res.statusCode} - https://${hostname}${path}`);
      let data = "";
      res.on("data", (chunk) => {
        data = data + chunk.toString();
      res.on("end", () => {
        const json = JSON.parse(data);
    req.on("error", (error) => {
 * Gets a JWT token for authorization.
 * @param {string} hostname
 * @param {string} path
 * @return {Promise<any>} a JWT token.
const getAccessTokenAsync = (hostname: string, path: string): Promise<any> => {
  const body = "grant_type=client_credentials";
  const authz = Buffer.from(`${CONSUMER_KEY}:${CONSUMER_SECRET}`).toString("base64");
  const headers = {
    "Content-Type": "application/x-www-form-urlencoded",
    "Cache-Control": "no-cache",
    "Authorization": "Basic " + authz,
    "Content-Length": body.length,
  return postAsync(hostname, path, headers, body);
export const getAccessToken = functions.https.onCall((_data, _context) => {
  // See:
  return getAccessTokenAsync("", "/v1/oauth2/token");

Download the source code of this Firebase Function.

Deploy this function to your Firebase project:

firebase deploy

Log into the Firebase Dashboard, navigate to your Firebase project, and select Functions in the Build menu.

After a few seconds you will see your getAccessToken function.

Note: The Memory is automatically set to 256 MB. Since the code is really small, you should be able to run it with only 128 MB. You can also lower the timeout. To change that, go to the Google Cloud console and update the function.

And that was it, we have deployed an authentication service using a serverless service on Google Cloud. Now, we need to modify our applications to leverage this authentication service and remove, once and for all, the Consumer Key and Consumer Secret from the source code of your applications. This method of initialization will be deprecated in a future release of the SDKs across all platforms.

Android implementation

The following code is based on the voxeet-sdk-android-gettingstarted GitHub repository.

In the file build.gradle, add the following classpath in the dependencies of the buildscript section:

classpath ''

In the file app/build.gradle, add the following dependencies:

// Import the BoM for the Firebase platform
implementation platform('')
// Declare the dependencies for the Firebase libraries
// When using the BoM, you don't specify versions in Firebase library dependencies
implementation ''
implementation ''

Download the google-services.json file from your Firebase dashboard and move it to the app folder of your Android project.

In your file, add the following imports:

import java.util.HashMap;

Create the following variable in the MainActivity class:

private FirebaseFunctions mFunctions;

Create the function that will request the Access Token from the getAccessToken* Firebase Function.

 * Get an Access Token from the Firebase Function.
private Task<String> getAccessToken() {
    return mFunctions
            .continueWith(task -> {
                HttpsCallableResult result = task.getResult();
                HashMap<String, String> json = (HashMap<String, String>) result.getData();
                return json.get("access_token");

In the onCreate(Bundle savedInstanceState) method, initialize the mFunctions variable, remove the deprecated initialize method, and request an access token from Firebase Function to initialize the SDK.

mFunctions = FirebaseFunctions.getInstance();
// throw new IllegalStateException("<---- Remove this line and set your keys below to use this sample !!");
// VoxeetSDK.initialize("", "");
        .addOnSuccessListener(accessToken -> {
            // Initialize the SDK with the Access Token
            VoxeetSDK.initialize(accessToken, (isExpired, callback) -> {
                // When the callback is triggered, request a new access token
                        .addOnSuccessListener(refreshToken -> {
                        .addOnFailureListener(error -> {
        .addOnFailureListener(error -> {

That’s it; the Android application is now using the Firebase Function to request an access token and initialize the SDK.

iOS implementation

The following code is based on the voxeet-sdk-ios-gettingstarted GitHub repository.

Download the GoogleService-Info.plist file from your Firebase dashboard and move it to your application folder.

Follow the instructions from the Firebase Documentation to install Firebase in your iOS application. But in summary, you will need to initialize CocoaPods with the following command:

pod init

Add the pod Firebase/Functions in the file PodFile. And install it with the following command:

pod install

A new .xcworkspace file will be created, open it with Xcode and open the file AppDelegate.swift. Add the following import:

import Firebase

Add the following code that will allow us to request a new access token from the Firebase function:

// Get the Firebase Functions
lazy var functions = Functions.functions()
private func getAccessToken(success: @escaping ((_ accessToken: String) -> Void), fail: @escaping ((_ error: NSError?) -> Void)) {
    // Request an Access Token
    functions.httpsCallable("getAccessToken").call() { result, error in
        guard let jwt = result?.data as? NSDictionary,
              let accessToken = jwt["access_token"] as? String,
              error == nil else {
            fail(error as NSError?)

Now, in the function application add the following code to initialize Firebase and the SDK:

// Use Firebase library to configure APIs
// Request an Access Token
getAccessToken { accessToken in
    // Voxeet SDK initialization
    VoxeetSDK.shared.initialize(accessToken: accessToken, refreshTokenClosureWithParam: { closure, isExpired in
        // Request a new Access Token
        self.getAccessToken { newAccessToken in
        } fail: { error in
} fail: { error in
    // Report an error.
//VoxeetSDK.shared.initialize(consumerKey: "YOUR_CONSUMER_KEY", consumerSecret: "YOUR_CONSUMER_SECRET")

That’s it; the iOS application is now using the Firebase Function to request an access token and initialize the SDK.

JavaScript implementation

The following code is based on the voxeet-sdk-browser-gettingstarted GitHub repository.

In the <head> of the index.htmlfile, add the following scripts to load the Firebase JavaScript SDK:

<!-- Google Firebase SDK -->
<script src=""></script>
<script src=""></script>

Go to the list of applications from your project and select the web application you want to configure.

Select the Config radio button and copy the firebaseConfig variable.

Open the client.js file and replace the try {} catch {} section with the following code:

// Your web app's Firebase configuration
// For Firebase JS SDK v7.20.0 and later, measurementId is optional
var firebaseConfig = {
  apiKey: "B0s4d5s64d65s4564ds56f4d6s54f6d5s4f-saddf5fdsfDF65ds",
  authDomain: "",
  databaseURL: "",
  projectId: "blog-post-01234",
  storageBucket: "",
  messagingSenderId: "000000000000",
  appId: "1:000000000000:web:0000011111222224444455",
  measurementId: "G-0000000000"
// Initialize Firebase
try {
  // Get the Firebase Function
  const getAccessToken = firebase.functions().httpsCallable('getAccessToken');
  // Request an access token
  const result = await getAccessToken();
  // Initialize the SDK
  VoxeetSDK.initializeToken(, async () => {
    // Refresh the Access Token
    const result2 = await getAccessToken();
  // Open a session for the user
  await{ name: randomName });
  // Initialize the UI
} catch (e) {
  alert('Something went wrong : ' + e);

That’s it; the JavaScript application is now using the Firebase Function to request an access token and initialize the SDK.

Tags: android, firebase, ios, javascript
How to Remove Background Noise From Video and Communicate Better as a Team

Background noise is distracting and makes it difficult for your team to work together effectively. Learn more about the impact of audio quality on collaboration and get some tips along the way.

Jessica Zhu
Best Practices for a Secure Video Conferencing Experience That Customers Trust

7 tips to make web conferencing more secure and develop customer trust
video conferencing
Recording Audio on iOS with Examples

Review of media frameworks available for recording audio on iOS.

Casper Mok
We're happy to chat about our APIs, SDKs...or magic.