Bear

Welcome to Bear Project

This project is currently in development phase.

The project aims to create a Cortana like application for Gnome 3 systems. The replies from Bear will be shown as cards in the material design. The following are the development phases (++ completed,-- not completed).

• ++ Card interface for displaying notification
• ++ Created D-BUS service for other applications to use the Card interface.
• ++ Api.ai, an online artificial intelligent agent's API calls
• ++ Voice Recognition engine defaults to Api.ai
• ++ Added Node.js based Google Speech API to enhance speech recognition accuracy
• ++ Created command line interface to call Bear
• ++ Created Makefile for easy installation.
• -- Display of pictures in Card interface
• -- Creation of a text field to type and search.
• -+ Error detection or log mechanism (Currently developing).

The project has only been tested in Fedora systems.

Screenshots

Main Panel Notification

Card pop-up Notification

Installation

1. On the extracted folder, type the command "make "

Pitfalls

If after installation Bear fails to run correctly, try these following measures:

1. Use Alt + F2 and type r to restart GNOME.
2. Make sure that there is microphone in your system and is not MUTED.

Google Speech API

To have a highly accurate prediction of voice (even during noise), use GOOGLE SPEECH API. To use Google Speech API, you need an API key from Google. You can obtain the key by following the steps in this Google API link.

Once you have obtained the key, few changes are needed.

1. In Bear python file, change using_google_speech_api = False to using_google_speech_api = True

2. In the .Bear/index.js file, add the api key value instead of GOOGLE SPEECH API KEY key: 'GOOGLE SPEECH API KEY' to like key: 'qwerty1234567890asdf'

How to run Bear

Speak to Bear after clicking speech icon at the main panel .

Say ,

         " What is Google?" or " Who is Bill Gates?" or "how are you"

to talk to Bear/api.ai engine. It records the audio for 10 seconds and sents it for processing to Api.ai speech recognition engine. If Google Speech API is enabled, then audio file is sent to Google Speech engine. After obtaining the audio text, it is sent for processing at Api.ai and results are displayed as card notification.

Also in terminal, you can call Bear command by giving

                   "Bear What is Google?"
                   "Bear Who is Bill Gates?"

Support or Contact

Having trouble with Bear? Mail me ( harishanand95 .at. gmail .dot. com) and we’ll help you sort it out.

Images from http://ic8.link/1893 ,http://ic8.link/2671