Speech to Text Design Project

This is a project made by a student group from McGill University as a design course project. Team member: Robyn Chen, Fei Feng, Yan Ren, Even Wang

This project would not have been possible without the help and kind support of our supervisors and the Radio-Canada Digital R&D Lab team. We would like to express our deepest appreciation to Xavier K. Richard and Thomas Le Jouan for their guidance and supervision as well as for providing support throughout the entire part one of this design project.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

What things you need to install the software and how to install them

• MongoDB - The backend database used
• NodeJS - The web framework used

Database setup

In app.js, find the code

mongoose.connect('mongodb://localhost:27017/test', { useMongoClient: true });

Replace with the database you are using.

Create two collections:

• media
• subs

Server setup

Install the dependencies

npm install

Start the backend

node app.js

Google Cloud Platform setup

Create Google Cloud Platform account and activate Cloud Storage and Cloud Speech API.

After Google Cloud Platform account, store the JSON key file in project root directory.

In /controller/storage.js, find the code

const bucketName = 'speech-to-text-sandbox';
const keyFileLocation = path.join(__dirname, '..', 'speech-to-text-sandbox-9b4c51ccdb39.json');

Replace with your JSON key file and cloud storage bucket.

In /controller/stt.js, find the code

const keyFileLocation = path.join(__dirname, '..', 'speech-to-text-sandbox-9b4c51ccdb39.json');

Replace with you JSON key file.

Test the project

Use .flac mono channel file to test the project. Or use the file in /testingAudioFile for simple testing.

https://cloud.google.com/speech/reference/rest/v1/RecognitionConfig#AudioEncoding