There's a small workaround for it. You can store the timestamp in a variable upon activating SpeechRecognition and when it exits after a few seconds of inactivity, it will be compared to a timestamp since the SpeechRecognition was activated. Since two tabs are using the API simultaneously, it will exit immediately.
For Chrome, you can use the code below and modify it base on your needs. Firefox doesn't support this yet at the moment.
var transcriptionStartTime;
var timeSinceLastStart;
function configureRecognition(){
var webkitSpeechRecognition = window.webkitSpeechRecognition || window.SpeechRecognition;
if ('webkitSpeechRecognition' in window) {
recognition = new webkitSpeechRecognition();
recognition.continuous = true;
recognition.interimResults = true;
recognition.lang = "en-US";
recognition.onend = function() {
timeSinceLastStart = new Date().getTime() - transcriptionStartTime;
if (timeSinceLastStart < 100) {
alert('Speech recognition failed to start. Please close the tab that is currently using it.');
}
}
}
}
See browser compatibility here: https://developer.mozilla.org/en-US/docs/Web/API/SpeechRecognition