dongido001 / call-to-text

Live Transcribing Phone Calls using Twilio Media Streams and Google Speech-to-Text

With Twilio Media Streams, you can now extend the capabilities of your Twilio-powered voice application with real time access to the raw audio stream of phone calls. For example, we can build tools that transcribe the speech from a phone call live into a browser window, run sentiment analysis of the speech on a phone call or even use voice biometrics to identify individuals.

Blog Post

If you prefer a step by step guide through building this yourself, this blog post will guide you through transcribing speech from a phone call into text, live in the browser using Twilio and Google Speech-to-Text using Node.js.

Visit the Blog Post here

---

Prerequisites

Before we can get started, you’ll need to make sure to have:

• A Free Twilio Account
• A Google Cloud Account
• Installed ngrok
• Installed the Twilio CLI

---

Setup

1. Setup Google Project and retrieve service account key

   a. Install and initialize the Cloud SDK

   b. Setup a new GCP Project

   c. Enable the Google Speech-To-Text API for that project

   d. Create a service account.

   e. Download a private key as JSON.

2. Modify the .env.sample file to include the path to your JSON service account key and save it as a .env file

3. Run the following commands:

   Buy a Phone Number (I have used the GB country code to buy a mobile number, but feel free to change this for a number local to you.)

   $ twilio phone-numbers:buy:mobile --country-code GB

   Start ngrok:

   $ ngrok http 8080

   While this is running in a new window copy the forwarding HTTPS URL (https://xxxxx.ngrok.io) and set your Twilio number to this URL:

   $ twilio phone-numbers:update TWILIO_NUMBER --voice-url https://xxxxxxxx.ngrok.io

   Install dependencies and start your server:

   $ npm install

   $ npm start