Our project A.V.E.R.I., which stands for Another Voice Expert Recognition Interface, will be a Google Chrome extension that further enhances Google’s voice recognition capabilities. The motivation for the project is J.A.R.V.I.S. from Marvel. While several voice assistants already exist in the market, we want to improve on their capabilities. For example, Google Assistant is a voice-controlled “personal assistant” that can perform various tasks for users - it can find information online, send messages, access information from calendars, set reminders, and several other features. With A.V.E.R.I., we want to build on these existing features by creating a voice recognition functionality for personalized, situation-based reminders and tasks. Members of the team have experience with C++ and are familiar with Python, HTML, and CSS.
The primary customer is anyone who does not have any prior knowledge on technology or anyone who may need additional accessibility to perform commands. The customer would want something that makes it easier to perform commands without having to manually do them or an ecosystem that is similar to Apple’s. The underlying problem is that many customers that were born prior to the technological boom of the early 2000’s or people with disabilities have limited knowledge of how to use technology on an efficient level, so allowing them to perform practical commands with their voice can bridge the gap between their skill levels. In the context of the market, there is a lot of software with similar functionalities using voice, but we aim to create an extension that can be used with Google software regardless of platform. Since we intend for this to be a Google extension, the Google Extension Store will serve as the media for our customers to write their reviews on. Generally, our customer-centric measures of success will depend on how many users deem our extension as essential to their everyday life.
Our project will deliver a convenient way to go about everyday life and tasks, starting with to do lists and reminders for recurring events. The vision is to be able to activate the voice recognition service to do simple tasks seamlessly, without ever looking down at your phone for every task.
Our software will listen to input from the user and do the intended actions that are personalized by the user. The main goal is to have a Google Chrome Extension that can be accessed through your phone and computer. It will store any input, mainly in the form of a list. When the extension is told that a specific task is happening, keyed on a timely reminder or a situation, the extension will respond with a personalized list of actionable tasks. For example, if you go on sporadic runs throughout the week, a reminder is not the most ideal. You still want to be reminded to bring a water bottle and earbuds. The extension will be able to take the input (i.e. “I’m going on a run now.”) and reply with the given list or reminders (i.e. “The weather is sunny at 60 Fahrenheit today. Remember to bring a water bottle to stay hydrated and earbuds to join the run”). It can have additional features like reminding you when sundown is if you go on a late night run. The extension will provide templates for possible given events, but it is mostly focused on how one can personalize it to individuals.
Testing includes making sure that keywords can be picked up by the extension and understood with colloquial language. The agile process will be used, with many iterations that continuously enhance the extension. The extension will be used by the customer for beta testing. We would like to add features that are requested by the customer, while further developing a library of capabilities for the extension.
We will use the guidelines from Chrome Developers to build the system with HTML and CSS for front end processes. For backend, Python, Javascript, and C++ will be used. Google Developers already has a Voice Driven Web Application that we intend to use for input.
None of the members of the team have built something like this before. All members of the team have proficiency in C++ and are familiar with Python, HTML, CSS, as well as back-end development. This will be a new experience for every member.
Every team member will have an equal part to play in the project. We have designated preliminary roles for each team member. The roles for this project will constantly be rotating as each team member looks to gain experience in multiple areas of the project.
Sehee - Back-End
Pei - Front-End and User Interface and User Experience
Andy - Back-End
Kedar - Front-End and User Interface and User Experience
| Week | Goals |
|---|---|
| 02/07/21 - 02/13/21 | - Create rough project proposal |
| 02/14/21 - 02/20/21 | - Complete project proposal and submit - Read documentations on Google Chrome Extensions - Begin creating basic extension |
| 02/21/21 - 02/27/21 | - Complete basic extension - Brainstorm basic features needed |
| 02/28/21 - 03/06/21 | - Submit Iteration 1 Status - Finalize features and begin building - Start on back-end development |
| 03/07/21 - 03/13/21 | - Begin Front-End development - Brainstorm UI/UX themes and structure |
| 03/14/21 - 03/20/21 | - Submit Iteration 2 Status - Work on Front-End and back-End |
| 03/21/21 - 03/27/21 | - Front-End and Back-End Development |
| 03/28/21 - 04/03/21 | - Submit Iteration 3 Status - Finalization and review |
| 04/04/21 - 04/10/21 | - Final testing and design changes - Prepare for presentation - Begin final report |
| 04/11/21 - 04/17/21 | - Present Final Project - Finish and complete final report |
While we intend to stay clear of any legal constraints or ethical concerns, the idea of a voice recognition software will cause people to be concerned with the idea that the software is listening to their everyday conversations.
We should have access to the data we need to create a working voice recognition Chrome extension that can perform commands with other software.
If the full functionality can not be implemented in time, we hope that our code will provide a sufficient basis for other developers to expand the capabilities of Google’s voice recognition software. The reduced functionality will operate as a simple voice recognition software that can handle creating reminders and pulling up hyperlinks. We believe that our software will provide a beginning to an ecosystem that rivals Apple.