0.9 release?

[Didier Spaier]

On 03/11/2018 22:11, Didier Spaier wrote:
> Nothing that I can thing of that absolutely needs to be done before 0.9.

But include the suggestions below in TODO if you agree ;)

> possibly after 0.9:
> - Allow setting a synthesis voice in the user config using spd-conf. I could
>   propose an implementation, occasion to learn a bit of Python. Caveat emptor:
>   this would make the user config include more parameters than the system one
>   as in my opinion this setting is not relevant system wide.
> - Get rid of the symbolic voices. It should be enough to state the gender. Of
>   course a transition should be provided, but just grep -i male or female in 
> the
>   client's command could be enough for now, the config files being just edited
>   to keep only the gender M of F.
> - As already suggested, make the variant independent of the voice name.
> - Make sure that in no circumstance the client be not provided a symbolic 
> voice.
>   If at least one is installed, of course ;)
>   Issues can happen for instance if a client or default setting (be it lang,
>   gender, synthesizer or synthesis voice) can't be honored, possibly because
>   there is no match, or there was one but some component (synthesizer or 
> voice)
>   has been removed since the setting has been done.
>   Ideally the server should always propose a voice, even if it doesn't meet 
> the
>   requirements it receives, possibly with a warning.
>   I realize that this would not be simple to implement, so should be 
> discussed 
>   by whoever in concern (developers of speech-dispatcher but also of clients).
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0xD50202EF60C03EEA.asc
Type: application/pgp-keys
Size: 2153 bytes
Desc: not available
URL: 
<http://lists.freebsoft.org/pipermail/speechd/attachments/20181103/17257d66/attachment.key>