Compiling Speakup modules

[Tony Baechler]

Hi all,

I've just been doing some poking around in the kernel sources and I see that
Speakup is in staging. A mention of this on the web site would be helpful
since I thought, based on what the site says, that it's still in the old git
repository and was disappointed to see that none of the code had been
updated in years. Anyway, someone is having a problem with the DECtalk
dropping pitch randomly. While I can't reproduce it here, I see that
speakup_dectlk.c has a line to drop the pitch to 100, so presumably changing
it to 120 would fix the problem.

My question is this: Once I make the change, how do I recompile the Speakup
modules? Do I have to compile a new kernel from scratch? I really don't
want to build a new kernel just for a single line change. If I can't just
build Speakup, can I only build the staging drivers? I see a Makefile in
the Speakup source, but it looks like it won't work by itself. I read the
README in staging, but it talked about compiling everything. I am not a
programmer and I'm not sure where to look. The kernel README said to look
at the Changes file, but that isn't in the staging tree. I fear I have to
download a full kernel release and do a full compile. Is this correct?

Finally, how do you get hardware speech working on kernels after 2.6.32? I
know of someone running 3.16 and he says it works, but I just installed 4.0
from Debian testing and it appeared to lock up the machine when I loaded the
speakup_dectlk module. This is the linux-image-686 metapackage. How can I
get around this and use official Debian kernels?

Thanks very much for your help.

--------------------
Tony Baechler, Baechler Access Technology Services
Putting accessibility at the forefront of technology
mailto:***@batsupport.com
Phone: [phone] Fax: [phone]

[Willem van der Walt]

No, you can just build the modules.
If some one else do not give you the propper make and make install
commands for this, I'll try to help, but will also have to read up a bit.

Post by Tony Baechler
Hi all,
I've just been doing some poking around in the kernel sources and I see that
Speakup is in staging. A mention of this on the web site would be helpful
since I thought, based on what the site says, that it's still in the old git
repository and was disappointed to see that none of the code had been
updated in years. Anyway, someone is having a problem with the DECtalk
dropping pitch randomly. While I can't reproduce it here, I see that
speakup_dectlk.c has a line to drop the pitch to 100, so presumably changing
it to 120 would fix the problem.
My question is this: Once I make the change, how do I recompile the Speakup
modules? Do I have to compile a new kernel from scratch? I really don't
want to build a new kernel just for a single line change. If I can't just
build Speakup, can I only build the staging drivers? I see a Makefile in
the Speakup source, but it looks like it won't work by itself. I read the
README in staging, but it talked about compiling everything. I am not a
programmer and I'm not sure where to look. The kernel README said to look
at the Changes file, but that isn't in the staging tree. I fear I have to
download a full kernel release and do a full compile. Is this correct?
Finally, how do you get hardware speech working on kernels after 2.6.32? I
know of someone running 3.16 and he says it works, but I just installed 4.0
from Debian testing and it appeared to lock up the machine when I loaded the
speakup_dectlk module. This is the linux-image-686 metapackage. How can I
get around this and use official Debian kernels?
Thanks very much for your help.
--------------------
Tony Baechler, Baechler Access Technology Services
Putting accessibility at the forefront of technology
Phone: [phone] Fax: [phone]
_______________________________________________
Speakup mailing list
http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup
--
This message is subject to the CSIR's copyright terms and conditions, e-mail
legal notice, and implemented Open Document Format (ODF) standard.
The full disclaimer details can be found at
http://www.csir.co.za/disclaimer.html.
This message has been scanned for viruses and dangerous content by
MailScanner,
and is believed to be clean.
Please consider the environment before printing this email.

[John G Heim]

I was interested in this question because there is the other bug where
speakup cannot talk to most serial hardware speech synthesizers. It
would be nice to be able to patch, compile, and install just the one
module that causes that. But googling for info on compiling and
installing a single module seems to indicate that it's no easier
(probably harder) than compiling the whole kernel.

Depending on your distro, recompiling the kernel isn't too hard. I have
a script for ubuntu that does the whole thing from downloading the
source to installing my patch to compiling the kernel. It's routine. It
does take a long time but you don't have to sit there and watch it the
whole time.

You should submit a bug report though so eventually, everyone can
benefit from your patch.

Post by Tony Baechler
Hi all,
I've just been doing some poking around in the kernel sources and I see
that Speakup is in staging. A mention of this on the web site would be
helpful since I thought, based on what the site says, that it's still in
the old git repository and was disappointed to see that none of the code
had been updated in years. Anyway, someone is having a problem with the
DECtalk dropping pitch randomly. While I can't reproduce it here, I see
that speakup_dectlk.c has a line to drop the pitch to 100, so presumably
changing it to 120 would fix the problem.
My question is this: Once I make the change, how do I recompile the
Speakup modules? Do I have to compile a new kernel from scratch? I
really don't want to build a new kernel just for a single line change.
If I can't just build Speakup, can I only build the staging drivers? I
see a Makefile in the Speakup source, but it looks like it won't work by
itself. I read the README in staging, but it talked about compiling
everything. I am not a programmer and I'm not sure where to look. The
kernel README said to look at the Changes file, but that isn't in the
staging tree. I fear I have to download a full kernel release and do a
full compile. Is this correct?
Finally, how do you get hardware speech working on kernels after
2.6.32? I know of someone running 3.16 and he says it works, but I just
installed 4.0 from Debian testing and it appeared to lock up the machine
when I loaded the speakup_dectlk module. This is the linux-image-686
metapackage. How can I get around this and use official Debian kernels?
Thanks very much for your help.
--------------------
Tony Baechler, Baechler Access Technology Services
Putting accessibility at the forefront of technology
Phone: [phone] Fax: [phone]
_______________________________________________
Speakup mailing list
http://linux-speakup.org/cgi-bin/mailman/listinfo/speakup

[Hart Larry]

Well John, in an off-list discussion, I sent Tony a portion of a message in this
list from February 01 of this year, where Andy Spensor, 1 of our LUG members
figured out a work around so I could run this later kernel. At the time they
were no other comments, as it seems maybe only 5 of us are running Dectalks with
Speakup? Hope that helps
Hart

[Tony Baechler]

The problem is that I'm not a programmer and I really don't have the
infrastructure or resources to compile a custom kernel. I have to be able
to test the patch before submitting it, but I really have no idea what I'm
doing. Now he says the pitch, volume and rate drop. I thought it was just
the pitch, so it looks like my simple fix won't work after all. On my main
machine, it takes at least two hours to compile and that's a bit long to
wait for a simple test. I have a server, but I access it remotely and I
would have to build, download and install kernel packages every time which
also isn't practical. That being said, the server runs Ubuntu 14.04, so I
would like to try your script. I know Debian has make-kpkg, but that in
itself wouldn't help since I have no idea how to write a patch which would
actually fix the problem. If I do get the problem fixed, I'll of course
post here.

Post by John G Heim
I was interested in this question because there is the other bug where
speakup cannot talk to most serial hardware speech synthesizers. It would be
nice to be able to patch, compile, and install just the one module that
causes that. But googling for info on compiling and installing a single
module seems to indicate that it's no easier (probably harder) than
compiling the whole kernel.
Depending on your distro, recompiling the kernel isn't too hard. I have a
script for ubuntu that does the whole thing from downloading the source to
installing my patch to compiling the kernel. It's routine. It does take a
long time but you don't have to sit there and watch it the whole time.
You should submit a bug report though so eventually, everyone can benefit
from your patch.