Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Intent to ship: Speech synthesis API

60 views
Skip to first unread message

Eitan Isaacson

unread,
Apr 28, 2016, 3:52:11 PM4/28/16
to dev-pl...@lists.mozilla.org
Hello,

In Firefox 49 I intend to turn on the speech synthesis API on by default in
desktop.

For those paying close attention, this is not the first mail like this.
Last time I turned it on in Nightly only to allow some fuzz testing. As of
this writing there is one blocker we are tracking. All the details can be
seen in bug 1268633.

Blink and Webkit both support synth in their stable releases.

Cheers,
Eitan.

Gijs Kruitbosch

unread,
Apr 29, 2016, 5:06:40 AM4/29/16
to Eitan Isaacson
There are still quite a number of bugs on file for this feature:

https://bugzilla.mozilla.org/buglist.cgi?quicksearch=speech%20synthesis&list_id=12990930

one of which seems to suggest we want to switch to a different system
API for synthesis on Windows. Are we sure none of these other bugs need
to block shipping this?

~ Gijs

Gregory Szorc

unread,
Apr 29, 2016, 10:54:40 AM4/29/16
to Eitan Isaacson, dev-platform
On Thu, Apr 28, 2016 at 12:52 PM, Eitan Isaacson <eisa...@mozilla.com>
wrote:

> Hello,
>
> In Firefox 49 I intend to turn on the speech synthesis API on by default in
> desktop.
>
> For those paying close attention, this is not the first mail like this.
> Last time I turned it on in Nightly only to allow some fuzz testing. As of
> this writing there is one blocker we are tracking. All the details can be
> seen in bug 1268633.
>
> Blink and Webkit both support synth in their stable releases.
>

Are we shipping extra data files to enable speech synthesis? If so, how
large are they? If so, why aren't we using the speech synthesis APIs built
into the underlying OS (especially on Windows, which is where most of our
users are)?

If not and the only overhead to this is the new compiled code, then great!

Eitan Isaacson

unread,
Apr 29, 2016, 4:25:02 PM4/29/16
to Gregory Szorc, dev-platform
>
> Are we shipping extra data files to enable speech synthesis? If so, how
> large are they? If so, why aren't we using the speech synthesis APIs built
> into the underlying OS (especially on Windows, which is where most of our
> users are)?
>
>
We are doing the latter (relying on platform tts support). No extra
libraries/data.


> If not and the only overhead to this is the new compiled code, then great!
>

Great, then!
0 new messages