Online Timbre Dissimilarity

A web-app for conducting an online timbre dissimilarity study, following the methodology set out in [1]. Built around lab.js and Express

Installation

Installation is straightforward with:

yarn install

Usage

The Node.js process can be started with:

yarn start

The app will listen on the port specified in the environment variable process.env.PORT. If not found, it will default to port 8081. The app will connect to the database specified in the environment variable process.env.MONGODB_URI. If not found, it will attempt to connect to the database test on localhost:27017.

Once started, the experiment can be run by simply navigating to the server address in a browser.

Accessing Data

Two API endpoints are provided for accessing auto-generated CSVs containing experiment data:

• /data/dissimilarity_scores.csv
• /data/questionnaire_responses.csv

Entries in these files are linked by the anonymous specId field which is a unique hash connected only to the experiment spec (sequence of trials) stored in the database. No personally identifying information is stored.

To-do

• Set up ESLint CI action
• Write retroactive unit tests
• Add testing to CI workflow
• Refactor dissimilarity screen code, breaking out components usable in other audio studies (pairwise audio player etc.)
• Refactor server.js
• Dockerize

References

[1] Zacharakis, A., Pastiadis, K., & Reiss, J. D. (2015). An Interlanguage Unification of Musical Timbre: Bridging Semantic, Perceptual, and Acoustic Dimensions. Music Perception: An Interdisciplinary Journal, 32(4), 394–412. https://doi.org/10.1525/mp.2015.32.4.394