speech-commands

Introduction

speech-commands is a simple attempt to convert a user (programmer) speech into phrase for text editor.

How it works?

1. After start atom-plugin grabs list of words from all opened text editors and sends it for listener.
2. listener init a voice recording process and after it gets voice command send it to Google Cloud.
3. After listerer gets parsed text it tries to replace signs and guess correct words basing on context.
4. listener sends response to atom-plugin with its guess, plugin inserts it into current cursor position.

Setup

• for atom-plugin: you'll need to npm install inside its directory and add plugin into Atom editor
• for listener: npm install and put file named gcloud.pass.json which contains API key for your Google Cloud account

Note

This is a VERY primitive approach for the difficult topic of speech recognition and translation. If you want to learn more or you are just an enthusiast, please feel free to e-mail me.