--000000000000346c0c059319c0e9 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Hi Gustav, I see, no worries. In okular/qtspeech news the other day I played with okular's speak current page/document a bit on linux here and found a couple of issues I'm considering how to fix: 1) QtSpeech on linux loads the speech plugin randomly unless a specific plugin is specified. Okular's use of QtSpeech doesn't specify a plugin which is probably best, but because of this if you have both the speech-dispatcher and the flite qtspeech plugin installed it randomly uses one or the other to speak. Here I don't have flite installed, but do have a working speech-dispatcher, so half of the time I wouldn't get any speech at all (when it chose the flite plugin). I'm thinking to fix this either okular needs a setting to let the user choose which speech plugin to use, or QtSpeech itself needs a configuration to let the user say which plugin they prefer. 2) There's not a way to pause/resume. For this I added a patch to okular that is up for review here: https://phabricator.kde.org/D24102 which seems to work nicely so far (when QtSpeech randomly chooses the right plugin anyway) to let me pause/resume what it's reading. As for general speak clipboard contents I think a good place to add this feature would be in klipper itself that comes with plasma-desktop. I'll get that in place next I think. My idea is to have a context menu item in Klipper's system tray icon to speak clipboard contents and also an associated action that could be bound to a keyboard shortcut so ctrl-c to copy, then whatever you assign to speak clipboard contents to speak what was copied, similar to how it's done with Jovie. As for hearing multiple voices overlapping each other, I don't see how that could be happening, but in order to try to recreate it here could you send me your speech-dispatcher configuration files? Basically all files under ~/.config/speech-dispatcher/ thanks, Jeremy On Sat, Sep 21, 2019 at 6:34 PM Gustav Degreef wrote: > Looking back over it, I realize the question does not make sense. I am > just struggling to figure out how to use Qtspeech in a simple way. > Thanks, Gustav. > > On 9/18/19 10:43 PM, Jeremy Whiting wrote: > > Hi Gustav, > > > > On Wed, Sep 18, 2019 at 6:52 PM Gustav Degreef > > wrote: > .... > > > > Maybe this is a > > dumb question, but would it work with Qtspeech?=C3=82 > > > > > > Would what work with QtSpeech? I'm not sure what you're asking here. > > ktts was renamed to jovie, neither of them uses QtSpeech. > > =C3=82 > > > > > > On 9/18/19 10:27 AM, Jeremy Whiting wrote: > > > ktts is what jovie came from. It's much much older and does the > same > > > thing (ktts just got renamed to jovie when I refactored ktts to u= se > > > speech-dispatcher instead of a bunch of it's own synthesizer > > modules). > > > > --000000000000346c0c059319c0e9 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Hi Gustav,

I see, no worries= . In okular/qtspeech news the other day I played with okular's speak cu= rrent page/document a bit on linux here and found a couple of issues I'= m considering how to fix:

1) QtSpeech on linux loa= ds the speech plugin randomly unless a specific plugin is specified. Okular= 's use of QtSpeech doesn't specify a plugin which is probably best,= but because of this if you have both the speech-dispatcher and the flite q= tspeech plugin installed it randomly uses one or the other to speak. Here I= don't have flite installed, but do have a working speech-dispatcher, s= o half of the time I wouldn't get any speech at all (when it chose the = flite plugin). I'm thinking to fix this either okular needs a setting t= o let the user choose which speech plugin to use, or QtSpeech itself needs = a configuration to let the user say which plugin they prefer.
2) = There's not a way to pause/resume. For this I added a patch to okular t= hat is up for review here: h= ttps://phabricator.kde.org/D24102 which seems to work nicely so far (wh= en QtSpeech randomly chooses the right plugin anyway) to let me pause/resum= e what it's reading.

As for general speak clip= board contents I think a good place to add this feature would be in klipper= itself that comes with plasma-desktop. I'll get that in place next I t= hink. My idea is to have a context menu item in Klipper's system tray i= con to speak clipboard contents and also an associated action that could be= bound to a keyboard shortcut so ctrl-c to copy, then whatever you assign t= o speak clipboard contents to speak what was copied, similar to how it'= s done with Jovie.

As for hearing multiple voices = overlapping each other, I don't see how that could be happening, but in= order to try to recreate it here could you send me your speech-dispatcher = configuration files? Basically all files under ~/.config/speech-dispatcher/=

thanks,
Jeremy

=
On Sat, Se= p 21, 2019 at 6:34 PM Gustav Degreef <gustav97@gmail.com> wrote:
Looking back over it, I realize the question does not= make sense.=C2=A0 I am
just struggling to figure out how to use Qtspeech in a simple way.=C2=A0 Thanks, Gustav.

On 9/18/19 10:43 PM, Jeremy Whiting wrote:
> Hi Gustav,
>
> On Wed, Sep 18, 2019 at 6:52 PM Gustav Degreef <gustav97@gmail.com
> <mailto:gus= tav97@gmail.com>> wrote:
....
>
>=C2=A0 =C2=A0 =C2=A0=C2=A0Maybe this is a
>=C2=A0 =C2=A0 =C2=A0dumb question, but would it work with Qtspeech?=C3= =82=C2=A0
>
>
> Would what work with QtSpeech? I'm not sure what you're asking= here.
> ktts was renamed to jovie, neither of them uses QtSpeech.
> =C3=82=C2=A0
>
>
>=C2=A0 =C2=A0 =C2=A0On 9/18/19 10:27 AM, Jeremy Whiting wrote:
>=C2=A0 =C2=A0 =C2=A0> ktts is what jovie came from. It's much mu= ch older and does the same
>=C2=A0 =C2=A0 =C2=A0> thing (ktts just got renamed to jovie when I r= efactored ktts to use
>=C2=A0 =C2=A0 =C2=A0> speech-dispatcher instead of a bunch of it'= ;s own synthesizer
>=C2=A0 =C2=A0 =C2=A0modules).
>

--000000000000346c0c059319c0e9--