Google voice AIY based chatbot
With the advancement in technology, virtual assistants such as Siri, Google Assistant and Alexa are on the rise and are widely used in the market now. The AIY Voice Kit is part of the AIY projects from Google which aims to help makers experiment with their own AI assistant. It utilizes the Google As...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2018
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/76162 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | With the advancement in technology, virtual assistants such as Siri, Google Assistant and Alexa are on the rise and are widely used in the market now. The AIY Voice Kit is part of the AIY projects from Google which aims to help makers experiment with their own AI assistant. It utilizes the Google Assistant to interact with users through voice. This project would explore how to build a virtual assistant using the Voice Kit and further improve on its performance by adding various functionalities.
The Little Ambassador was created to be the School of Computer Science and Engineering Office’s own virtual assistant. Located in the office’s main lobby, its purpose was to provide visitors with useful information. Therefore, a survey was conducted among a group of 42 students to find out which information is most valuable to them when they visit the office.
Subsequently, data was crawled from the school website to provide a database for Little Ambassador. The database was hosted on Firebase Realtime Database. With the aid of Google Assistant and Dialogflow, Little Ambassador is able to hold an interactive conversation with the user, providing multi-level interactions.
Additionally, the Little Ambassador is also equipped with a facial recognition module that was implemented with a PiCamera. It allows users to query staffs whom they do not know by showing a picture of their face and the information would be displayed on a user interface for the users to view. The user interface was constructed using the Django web framework and hosted locally on the Raspberry Pi.
Although Little Ambassador also made use of the Google Assistant SDK, there are some huge differences between this system and Google’s own products such as Google Home and Google Home Hub. An example is the ability to add in custom module such as the Facial Recognition module. Ultimately, being able to customise its functions had open Little Ambassador up to many opportunities for improvements in the future. |
---|