| Group |
2019-09 |
Status |
completed |
| Title |
mimik Capstone Project – 2 (Personal Assistant – Voice recognition |
| Supervisor |
W. Hamou-Lhadj |
| Description |
In the hyper-connected world, all things (nodes) need to communicate at the application level. An application (or a system) requires to compile/process data and/or knowledge from multiple sources (other applications and/or nodes). Voice will be a key interface with many systems. The main driver of the shift towards voice user interfaces is an increased overall awareness and a higher level of comfort demonstrated specifically by millennials. The mass adoption of artificial intelligence in users’ everyday lives is also fueling the shift towards voice applications. The number of IoT devices such as smart thermostats, appliances, and speakers are giving voice assistants more utility to connected users’ lives.
Consumers use voice assistants in specific locations, usually while multitasking, and can either be alone or amongst a group of people when using them. Conversational voice interface is nice and convenient and/or fun for most scenarios but is crucial when driving a car. In such environments, devices should be able to decipher the context such as noisy environments to make the conversation more convenient, but it is also crucial to minimize latency in communication and processing. Sending the data to central cloud for processing, driving knowledge, taking action and sending the response back may create a poor user experience and is obviously very inefficient. Imagine a voice activated hyper connected world where everything one says needs to first go to cloud for AI processing. This won’t be scalable in a cost-efficient manner and will suffer from low reliability and high latency. One must keep in mind that voice communication inside is not always limited to convenience and entertainment but could be about interaction with the car itself.
|
| Student Requirement |
- How deep learning works https://en.wikipedia.org/wiki/Deep_learning
- How to train a model for ML Inference
the example we offer is based on Caffe https://caffe.berkeleyvision.org
- How to work with RasberryPi https://www.raspberrypi.org/downloads/raspbian/
- How to work with RESTful API https://en.wikipedia.org/wiki/Representational_state_transfer
- How to work with JSON data format https://www.json.org
- How to work with Linux Ubuntu https://ubuntu.com
|
| Tools |
- - Raspberry Pi Hardwares
- Laptop/PC running Ubuntu 18.04 LTS
- Visual Studio Code as development IDE
|
| Number of Students |
3-5 |
| Students |
Jodoin,Filip
Tardieu,Victor
Anas Shakra |
| Comments: |
|
| Links: |
|