Re: Setting Default Voice for Module

[Didier Spaier]

Howdy Storm and All,

To set a default voice in speechd.conf in a safe way you would need to be sure
that this voice be available, but this depends on which other packages (TTS
engines and associated voices) are installed. What would happen if this voice is
not available in the system?

spd-say can set a specific voice as it scans available voices but you have to
set the module to use if not already done, like with these commands:

didier[~]$ spd-say -O
OUTPUT MODULES
pico
rhvoice
flite
espeak-ng-mbrola
espeak-ng


didier[~]$ spd-say -o pico -L
                     NAME                 LANGUAGE                  VARIANT
                 samantha                    en-US                     none
                   serena                    en-GB                     none
                  sabrina                    de-DE                     none
                   isabel                    es-ES                     none
                 virginie                    fr-FR                     none
                   silvia                    it-IT                     none

Then:
didier[~]$ spd-say -o pico -y samantha "Please repeat this."
or here:
didier[~]$ spd-say -o pico -y virginie "Répète-moi ça, s'il te plaît."

So to set a default voice you would need to also set accordingly the default
module in speechd.conf and hope that both are actually installed.

Incidentally some of the scripts in
https://github.com/DidierSpaier/slint-scripts could be useful for stormux, like
spd-list or list-espeak-ng-voices.

Cheers,
Didier


Le 2/13/24 à 22:38, Storm Dragon a écrit :
> Howdy All,
> 
> I was looking at some of the module files for voices. I wonder how to
> set the default voice for a module. The voxin one, for example talks
> about a table where all parameters are required, but the espeak-ng.conf
> file doesn't seem to have  a default voice parameter at all.
> 
> I tried adding:
> 
> DefaultVoice  Name
> 
> in the speechd.conf file but it doesn't seem to change anything. If I
> call spd-say -y name, though, it speaks with the specified voice. I
> guess my real question is what's the config file version of -y?
> 
> Thanks,
> Storm