[mary-users] making new voices

[Ingmar Steiner]

Hi Josh,

the voicebuilding for the more natural-sounding unit-selection voices is 
theoretically possible under Windows, but still untested. HMM-based 
voices, unfortunately, cannot be built without some modifications. Both 
voicebuilding variants rely on third-party tools and scripts, some of 
which are extremely tricky to get working on Windows.

But we're working on improving the voicebuilding in the coming months, 
so watch this space!

Best wishes,

-Ingmar

On 01.02.13 04:20, Josh Kennedy wrote:
> Hi
> do you absolutely need Linux to make new voices for openMary? can it be
> done under windows? do the voice building tools work with orca? if I
> wanted to make a unit selection voice of a favorite talking book
> narrator could I record the wav files and just type the text
> transcription of what is said in those files? and how many recordings of
> sentences would I need? I would like to make a tts voice if possible of
> jim dale, the guy who reads the harry potter books, and also of eric
> sandvold another talking book narrator.
> but I am not sure if the voice building tools are accessible. getting
> the line-in cable and USB sound card won't be an issue the issue is
> moreso accessibility of the tools. I'd rather do it with openmary
> because I hear festival is harder to make voices for.
> Josh
> _______________________________________________
> Mary-users mailing list
> Mary-users at dfki.de
> http://www.dfki.de/mailman/cgi-bin/listinfo/mary-users
>