Re: speechd-el issues / questions

[Otto Diesenbacher-Reinmüller]

Hi Milan,

Milan Zamazal <pdm@zamazal.org> writes:

>     ODR> * swallowing first char after entering mini-buffer - having set
> [...]
>     ODR> following characters are spoken correctly. This is rather
>     ODR> irritating.
> I can reproduce it, it is a bug.  I’ll look at it.
>     ODR> * 'C-e C-l'-spell-mode does not work spell mode isn't
>     ODR> working. 
> This has been reported by Cleverson recently, I plan to look at it too.

Many thanks - looking forward to your fix. Let me know if you need any
further information, testing, or any support.

>     ODR> * make a pause speaking I enjoy listening to EBooks opened in
>  [...]
>     ODR> line-endss get converted to a 1 second pause?
>
> I’m not aware about an easy way to do it.  If Speech Dispatcher doesn’t
> offer such a functionality, maybe it could be implemented in speechd-el,
> the question is how.  One way would be using SSML <break>, which
> requires support in the Speech Dispatcher output module.  Other options
> may be limited as it seems there is no SSIP command to insert a pause.
> If there is no better mechanism then a silent sound icon could do the
> job.

thanks, for the hint - for inserting a silent icon - I would also patch
/ advise speechd-speak--text? F.e. a dirty hack - setf some variable
make-breaks to t - and in speechd-speak--text substitute "\n\n" by the
icon if make-brekas is t? How should the insertion of the icon into the
text argument should look like? Or will I have to break up text argument
into consecutiove speechd-speak--text calls?

btw - what is the best source / docu to listen to, to understand
communication with speech-dispatcher better, beside speechd-el sources?

>     ODR> * filter or substitute characters Some texts that I listen to
> [...]
>     ODR> buffer-local?
>
> Not directly.  For those who can program in Elisp, it’s possible to
> advise speechd-speak--text function and transform its ‘text’ argument.
>
> What’s exactly your motivation for client-side transformations?

The PDFs, EPUBs, papers, journals, etc that I listen to have very
various unicode chars for f.e. quotes, or other delimiters, that
shouldn't be spoken, or rather substituted.  And in other documents,
these chars actually should be spoken, etc. So it varies... So I don't
want all the time edit config files, ask speech-dispatcher to reload
configs, etc.

But I will probably make a hack with speechd-speak--text function,
takening care of above mentioned pauses and substitution.

>     ODR> my speechd-el specific Emacs config, see
>     ODR> https://diesenbacher.net/blog/entries/Updated-speechd-el-config.html
>     ODR> perhaps also of interest for other speechd-el users.
> Cool, you’re quite advanced speechd-el user!

Thanks :)

> Regards,
> Milan

br okflo

-- 
Otto Karl Florian Diesenbacher-Reinmüller / okflo
okflo@diesenbacher.net
http://diesenbacher.net