Erie (Web)

Erie for Web (or simply Erie.js) is a web implementation of Erie, a declarative grammar for data sonification. Erie.js is intended for supporting sonification, not as auxiliary for visualization.

What Erie.js supports

• Writing a sonification specification
• Playing a sonification

What Erie.js does NOT do

• Making a “well-designed/good” sonification → It’s your job to do!

Some notes

Erie is looking for your input

Erie is an open-source project, and your input (bug report, code suggestion, etc.) is valuable. Currently, Erie is maintained by a single person (me), so potential collaboration is much welcomed. There are several ways you can contribute:

1. By reporting bug in the package repo (link).
2. By emailing me to discuss with potential collaboration (hyeokkim777 ‘at’ gmail ‘dot’ com).
3. Take a look at the bottom of this page for future extension plans.

Open-source projects and standard APIs used in Erie

Erie for Web uses W3C’s standard APIs for audio: Web Audio API and Web Speech API. For URL-based imports, Erie uses the standard Fetch API.

Many parts of Erie relies on Vega/Vega-Lite and D3. (Thousands of thousands of Thank you for them).

Precision of timing

In most cases, tones are timed/scheduled to be played as indicated. However, the timing of resulting sonification may not be of high precision in some environments when there are too many tones to play at the same time. For example, if you use a discrete tone for more than 100 data points withint too short time, then the Web Audio API can be overloaded, so it fails to play such sound (it might cause your browsers to be shut down). In that case, use a continuous tone.

Recording

The Web APIs do not support directly capturing audio generated from your browser or desktop (presumably for security, privacy, and copyright reasons). Instead, we use a Chrome extension by utilizing Chrome’s tabCapture API to generate audio files out of Erie sonifications (access here). Yet, it is still not possible to capture the sounds generated using the Web Speech API (SpeechSynthesis) because technically it is not generated from your browser but from your computer. Instead, our recorder creates an HTML file that has the text parts and recorded sounds so that they can be played using a screen reader.

Offline audio generation

Now, Erie supports offline audio generation. Refer to this documentation. Note that you need a separate Text-to-Speech API (usually a commercial service). To better support, Erie offers an interface function for Google Cloud Text-To-Speech API. In case you don’t have an access to a TTS API, you can still use the recorder extension.

Future expansion plans

• Support for 3d pan encoding