Skip to content

Instantly share code, notes, and snippets.

@wesbos
Last active July 20, 2022 18:13
Show Gist options
  • Save wesbos/cd16b8b1815825f111a2 to your computer and use it in GitHub Desktop.
Save wesbos/cd16b8b1815825f111a2 to your computer and use it in GitHub Desktop.
// paste in your console
speechSynthesis.onvoiceschanged = function() {
var msg = new SpeechSynthesisUtterance();
msg.voice = this.getVoices().filter(v => v.name == 'Cellos')[0];
msg.text = Object.keys(window).join(' ');
this.speak(msg);
};
@mrspeaker
Copy link

Damn you! Just redundified my js1k entry!

@jnewc
Copy link

jnewc commented Feb 17, 2016

"Good News" 👍

@chrislaughlin
Copy link

👍

@LPGhatguy
Copy link

I was able to get all the voices to work on OS X with Chrome 49, but on Windows with Chrome 50 I'm not getting any of the named voices, just the boring Google ones. 😦

@L1fescape
Copy link

Awesome! You can also invoke speech synthesis from outside the onvoiceschanged method:

var msg = new SpeechSynthesisUtterance();
msg.voice = speechSynthesis.getVoices().filter(v => v.name == 'Alex')[0];
msg.text = 'hi there from the dev tools console';
speechSynthesis.speak(msg);

@StoneCypher
Copy link

In case you'd like to know what they all sound like (there's a bunch of funny ones)

var msgs    = speechSynthesis.getVoices().map( (v, i) => { return { voice: v, text: `hi there from the voice named ${v.name}` }; }),
    speaker = new SpeechSynthesisUtterance(),
    halt    = false;

function speakOne(i) {
  if (halt) { halt = false; return; }
  speaker.voice = msgs[i].voice;
  speaker.text  = msgs[i].text;
  speechSynthesis.speak(speaker);
  console.log(`Saying "${speaker.voice.name}"`);
}

function speakAll() {
  var cursor = 0;
  speaker.onend = () => (++cursor) >= msgs.length? true : speakOne(cursor);  // chain each end
  speakOne(cursor);                                                          // start the series
}

To bail, set halt to true in the console. speechSynthesis.cancel() will only cancel the currently speaking voice.

I like Bad News, Bells, Cellos, Hysterical, and the way the language-specific Google voices butcher the words "hi there from."

Call speakOne(i) to listen to any specific voice by index, or speakAll() to listen to them all.

I also gisted this here.

@msikma
Copy link

msikma commented Feb 18, 2016

I've got Japanese too. Is this standard in all Chrome installs?

var msg = new SpeechSynthesisUtterance();
msg.voice = speechSynthesis.getVoices().filter(v => v.name == 'Google 日本語')[0];
msg.text = 'こんにちは皆さん!';
speechSynthesis.speak(msg);

@ToreJuloe
Copy link

@akenn The reason it's wrapped in the voiceschanged eventlistener, is that the voices are loaded asynchronously and this way you can make sure that they actually exist before you start using them.

It's no problem when you just run the snippet in the dev console, but if it's embedded on a page and set to run immediately, you'll get an error.

Here's a pretty cool Pen to play around with the different voices.

@L1fescape
Copy link

@ToreJuloe oh good to know - thanks!

@mathiasbynens
Copy link

Why not use Array#find? .filter(v => v.name == 'Cellos')[0].find(v => v.name == 'Cellos')

@msikma Not in my Opera/Chrome.

@riston
Copy link

riston commented Feb 19, 2016

Its not working in Linux window.speechSynthesis.getVoices() returns empty array no "voices" :(

@incleaf
Copy link

incleaf commented Mar 14, 2016

Wow 👍

@imtaehyun
Copy link

Love it!

@pokono
Copy link

pokono commented Mar 24, 2016

Awesome!!

@heytulsiprasad
Copy link

I really hope I wasn't the last one to find out this! Amazing 🎉

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment