On Sat, Apr 21, 2018 at 02:15 pm, Brian's Mail list account wrote:
Does this not then add one more parameter to change for programs like nvda?
Only if you want to change the base voice/accent. It will be like selecting Edward, or any of the other voice variants. And it would be no different to other TTS programs where you specify the voice and language to use.
Sent via blueyonder.
Please address personal E-mail to:-
briang1@..., putting 'Brian Gaff'
in the display name field.
----- Original Message -----
Sent: Saturday, April 21, 2018 4:31 PM
Subject: [espeak-ng] [espeak-ng:master] reported: Could you please re-add
phoneme equivalent tables? #github
[espeak-ng:master] New Comment on Issue #466 Could you please re-add phoneme
equivalent tables? ( https://github.com/espeak-ng/espeak-ng/issues/466 )
By rhdunn ( https://github.com/rhdunn ) :
Those tables were disabled, and only had code for compiling them in eSpeak.
As such, they did not work, and would not work if re-added.
I like the basic idea, but I want to do it in a more general/robust way.
Specifically, I want to split the speaker (voice) and their accent from the
language and it's accent. For example, you could have a Bavarian German
voice speaking in English with a Southern British English accent.
This would also apply to switched languages.
This is similar to #28 where I want to make the MBROLA voices into proper
voices. Here, instead of having mb-de5-en , you would specify mb-de5 as the
voice and en as the language/accent.