| 1 |
On Sat, 10 Oct 2020, Ashley Dixon wrote: |
| 2 |
|
| 3 |
> Date: Sat, 10 Oct 2020 08:17:07 |
| 4 |
> From: Ashley Dixon <ash@××××××××××.uk> |
| 5 |
> Reply-To: gentoo-user@l.g.o |
| 6 |
> To: gentoo-user@l.g.o |
| 7 |
> Subject: Re: [gentoo-user] |
| 8 |
> |
| 9 |
> On Sat, Oct 10, 2020 at 07:45:14AM -0400, Jude DaShiell wrote: |
| 10 |
> > I didn't emerge portaudio or pulseaudio before emerging espeak so will |
| 11 |
> > have to reemerge espeak to pick those USE variables up. |
| 12 |
> |
| 13 |
> You don't *need* portaudio or Pulse, but then you'll only be able to create WAV |
| 14 |
> files, and not have the audio played live [1]. The developers should probably |
| 15 |
> set one of them to be enabled by default in IUSE, since only creating WAV files |
| 16 |
> is a very unusual use-case for a screen-reader. |
| 17 |
> |
| 18 |
> > Another mistake I made was emerging espeak before emerging |
| 19 |
> > sys-kernel/gentoo-sources but since I'll have to reemerge, the |
| 20 |
> > sys-kernel/gentoo-sources package has been emerged on the system now. The |
| 21 |
> > right order of operations here is critical! |
| 22 |
> |
| 23 |
> That's quite rare for Gentoo; Portage usually takes care of all that type of |
| 24 |
> thing without requiring manual user interaction. The gentoo-sources ebuild |
| 25 |
> doesn't really do much, aside from calling a couple of functions in the |
| 26 |
> `kernel-2` eclass [2, 3] to extract the sources, generate the symlinks, and |
| 27 |
> check for any potential versioning issues. |
| 28 |
> |
| 29 |
> Can you provide some more details? Why is the order relevant? |
| 30 |
> |
| 31 |
> [1] https://gitweb.gentoo.org/repo/gentoo.git/tree/app-accessibility/espeak/espeak-1.48.04-r1.ebuild#n93 |
| 32 |
> [2] https://gitweb.gentoo.org/repo/gentoo.git/tree/eclass/kernel-2.eclass#n1603 |
| 33 |
> [3] https://gitweb.gentoo.org/repo/gentoo.git/tree/eclass/kernel-2.eclass#n1005 |
| 34 |
> |
| 35 |
> |
| 36 |
|
| 37 |
-- |
| 38 |
|
| 39 |
If you have portaudio or pulseaudio and alsa in your USE variables already |
| 40 |
then espeak will pull those in and build so it can do more than make wav |
| 41 |
files. Same with speech-dispatcher for other screen readers. I was being |
| 42 |
conservative with what I put in my USE variable and added things to it as |
| 43 |
I found things out. |
| 44 |
|
| 45 |
I did make menuconf in /usr/src/linux and in devices->staging drivers I |
| 46 |
found nothing to enable. Speakup got moved out of staging so that's |
| 47 |
understandable. in device drivers->accessibility all I found was enable |
| 48 |
app-accessibility which I turned on. Is speakup.synth=soft stored in the |
| 49 |
runtime driver for espeak now? |
Archives