This repo shows how to code an online speech recognition experiment using JavaScript and the JSpsych (https://www.jspsych.org/7.3/) library. Speech and text data recorded from user's browsers are saved in Dropbox until space is available. The experiment is deployed in Heroku. Link: https://articulatorycontroldemo.herokuapp.com/. However, link may break due to changes in Heroku's policies.