nkansal96 / aurora-reminders-app

A voice-enabled reminders application built with Aurora

Aurora Reminders App

Aurora Reminders is a voice-enabled reminders application built using Python and the Aurora SDK. Users can interact with a virtual chat assistant and use natural language to easily add reminders to their Google Calendar.

Features

• Transcribes recorded user speech into text
• Interprets natural language to create reminders in Google Calendar
• Responds to the user with synthesized speech

Getting Started

The Recommended Python version is 3.0+

1. Clone this repository.

2. Install dependencies with pip install -r requirements.txt.

3. Enable the Google Calendar API and copy your client_secret.json to the root directory of this project.

4. Create an account with Aurora.

5. Create a file named auroraKeys.py in the root directory.

APP_ID = YOUR_APP_ID
APP_TOKEN = YOUR_APP_TOKEN

6. Run the app using python3 speech.py.

Main Dependencies

Here are some of the core Python dependencies used in this project. The full list of requirements is available in requirements.txt

• Kivy
• Aurora Python SDK
• Google Client Library

Another requirement to get the audio working is ffmpeg. This can be downloaded here.

FAQ

1. I'm having trouble installing Kivy
Depending on what version of Kivy you are trying to install, it requires a different version of the dependency Cython. In our app, we used Kivy==1.10.0 and Cython==0.26.1. Furthermore, SDL2 should be installed as certain functions in the Kivy library require it. For further issues, reference the Kivy installation guide here.

2. I have Kivy installed properly, but I'm still getting an error as follows:

X Error of failed request:  BadWindow (invalid Window parameter)
  Major opcode of failed request:  4 (X_DestroyWindow)
  Resource id in failed request:  0x0
  Serial number of failed request:  160
  Current serial number in output stream:  162

What worked for us is to disable multisampling. This can be done by editing the file ~/.kivy/config.ini and setting multisamples to 0.

3. I got the following warning: UserWarning: Cannot access credentials.json: No such file or directory. How can I get rid of it?
No need to worry about this. The credentials.json is created when you first log in and caches your information into the directory. This speeds up future log ins. Just make sure client_secret.json is in your current directory as the credentials will eventually expire.