[LAU] Synthesized voices [was :Re: [LAD] Kontakt sampler format (and others like EXS24)]

[Patrick Shirkey]

On Mon, September 3, 2012 6:46 am, SxDx wrote:
> ----- Original Message -----
>> From: "Patrick Shirkey" <pshirkey at boosthardware.com>
>> To: linux-audio-user at lists.linuxaudio.org
>> Sent: Saturday, September 1, 2012 11:25:26 PM
>> Subject: [LAU] Synthesized voices [was :Re: [LAD] Kontakt sampler
format (and others like EXS24)]
>>
>>
>> On Sun, September 2, 2012 6:18 am, SxDx wrote:
>> > ----- Original Message -----
>> >> From: "Ralf Mardorf" <ralf.mardorf at alice-dsl.net>
>> >> The East West Choirs sing your lyrics. It's not another Ahhhh or Ohhhh
>> >> sample ;).
>> >>
>> >> http://www.youtube.com/watch?v=aI5Gg2-mhmU
>> >
>> > any technical info on that? free software doing the same, research
papers
>> > (easily available), anything?
>> >
>>
>> As an aside, one possible way of dealing with this issue is to create your
>> own voices for applications like orca.
>>
>> If you want a quick and easy way to get started with that process you can
>> run "AT Distro" in a virtual machine.
>>
>> http://atdistro.com
>
> thanks for pointing that out. I tried a bit, it's slow
> in qemu (expected). But it seems orca uses espeak, and
> the voice is not realistic.

You can create your own voices so the default voice should not be used as
the reference for a singing voice. Even then I was amazed by how close to
real it sounded for chanting emulation.


> What we hear in the video
> above is much more realistic and I wondered how they do
> that. Seems it's not public knowledge.
>

TTS is a huge research area.

Some people are putting their work online.

http://gitorious.org/lauloid

http://stephane.magnenat.net/music.html

IIRC there is some very advanced open source voice synth software
originating in France. I can't recall the name of the institute who
sponsors the development but I'm sure there are others on this list who
know of it.




--
Patrick Shirkey
Boost Hardware Ltd