Example: Voice Detection BrightSign Model Package (BSMP) Demo using an HTML5 Application

This demo HTML/JS application showcases the tech behind the NPU that is enabled in Brightsign players. This demo highlights:

• Full motion video
• Detects a person close enough to be heard by using a video model to "see" a person
• Listens for key words and triggers different videos

Key Words

• Chocolate
• Pizza
• Drinks
• Shampoo

Just Use It!

1. ensure you have all the pre-requisites per the BSMP
2. copy the contents of the 'sd' folder to an SD card
3. copy the BSMP onto the SD card
4. place the SD card into the player and boot

Prerequisites

This project assumes you are working in Linux. On MacOS we get an error in the dependencies:

npm error Error: Cannot find module 'node-bin-darwin-arm64/package.json'

You can also just use a Linux container to do the build steps. The container we often use is

• brightsign-dev-container

Building the App

First, clone the repository. Then, from the home directory of this repo:

make prep
make build

Deploying the Application

The easiest way to build the file folder you need is to:

make publish

This should place all the files you need into the "sd" folder. It should look like this:

sd
├── autorun.brs
└── dist
    ├── bundle.js
    ├── chacolate.mp4
    ├── drinks.mp4
    ├── index.html
    ├── meet-brightsign.mp4
    ├── pizza.mp4
    └── shampoo.mp4

Ensure the BSMP is Installed

The makefile automatically downloads and copies the BSMP to the sd folder. However, if you want to manually do this step:

• download the gaze detection bsfw installation package
• copy the file to the root of the SD card
• it will be automatically installed on the next boot

Licensing

This project is released under the terms of the Apache 2.0 License.