speakup using different synths with software speech?

Michael Whapples mwhapples at aim.com
Thu Jun 4 18:14:50 EDT 2009


Firstly orca and that dectalk express. You should be able to use it (I 
haven't tried it but when I installed emacspeak and had a look whether 
orca would support my apollo through the emacspeak speech server I 
noticed dectalk express in the list, my apollo wasn't). Anyway all I had 
to do was install orca, install emacspeak and emacspeak speech server 
(debian package emacspeak-ss) and then when I started orca preferences 
in the speech server combo box there was an option for emacspeak speech 
server and in the synthesiser combo box I could select dectalk express. 
I hope that works.

As for voxin, if you weren't pleased by eloquence, you are unlikely to 
be pleased with voxin as it uses IBM viavoice which sounds the same as 
eloquence (or at least very close, I think the voices are very slightly 
different).

As for contributing to espeak, you could suggest how it could sound 
better, it may not require altering the actual code it may be a matter 
of altering the voice files. I don't mean it in a bad way, but you seem 
to be hard to please with speech output, it may help if we knew what 
makes a voice good to you.

As for saying that software speech has a significant CPU overhead, it 
depends on what you use, realspeak is very slow to respond and seems to 
need a lot of processor to work, at the other end of the scale there is 
things like eloquence/viavoice and espeak which I feel are insignificant 
with a reasonable spec computer (I believe orca needs more processor 
power than espeak or viavoice, and when eloquence is used on mobile 
phones for talks and it is very useable it indicates it isn't heavy on 
resources).

Michael Whapples
On -10/01/37 20:59, Tony Baechler wrote:
> <div class="moz-text-flowed" style="font-family: -moz-fixed">Hi all,
>
> Apparently my position was unclear.  I'm aware of Voxin and I might 
> look at it, although I understand that it requires older libraries and 
> can be difficult to set up.  I'm on x86-64 and I don't know if there 
> is a 64-bit version.  I really don't mind paying for software speech 
> in principle.  The simple fact is that I don't like any software 
> speech, regardless of what it is.  I've used RealSpeak, Eloquence, 
> ESpeak, Festival, AT&T Flextalk, and the software DEC-talk.  I really 
> didn't like any of them, although some were tolerable.  They all have 
> a high memory and CPU overhead which hardware doesn't.
>
> I wouldn't be opposed to contributing somehow to ESpeak, but I really 
> don't have the money (see my previous post) and I'm not a programmer.  
> I'm really not sure what I could do.  It seems to work well enough, so 
> it isn't like there are bugs to report.  The documentation seems fine, 
> at least I had no problem using it with NVDA, Orca or ESpeakup.  I am 
> not trying to complain specifically about ESpeak, my comments 
> generally apply to all software speech.  I don't think I could get 
> used to the ESpeak voice for hours at a time though.
>
> Finally, if I did spend a fairly large amount of money on software, 
> whether it's speech or something else, I would want it to be open 
> source.  Yes, the GPL does allow companies to charge for software, as 
> long as the source is included.  I'll probably never use the source, 
> but at least I could recompile it on whatever system I'm using, such 
> as Debian Lenny on x86-64.  So, unless I'm mistaken, not only is the 
> software itself non-free, but they want money for it besides and it 
> still is not and never will be free software.  Surely people here can 
> understand why I would not want to use non-free software, regardless 
> of how much money is involved.  No, this really isn't about paying for 
> Windows, and only somewhat about the money involved.  Yes, call it my 
> prospectives or my principles, but I will not use software speech 
> under those conditions.  If it comes down to using non-free software 
> speech on Linux or using good hardware speech on Windows, since the 
> dealer installed Windows for me already and as such it doesn't cost me 
> anything, I'll keep using my old Windows 98 until it dies.  I can 
> afford to wait a long time for a free alternative for Linux to come 
> along.
>
> One last question on Orca and hardware speech.  I've used the DEC 
> Express with Emacspeak before, so I know it works and is supported.  I 
> don't know anything about speech servers though.  Would Orca support 
> that?  What about the Doubletalk LT, which I think is also supported 
> by Emacspeak?
>
> Michael Whapples wrote:
>> Firstly orca and hardware synths:
>> I think some are supported via emacspeak speech servers. I don't know 
>> how well this works and I believe it is limited to certain synths.
>>
>> As for free software speech:
>> I have to say it is a bit of either take what is there (eg. espeak) 
>> or pay your money for better. Voxin as I remember don't charge a huge 
>> amount, it was about 5 euros when I bought it, really not much if the 
>> quality of the speech is so important to you.
>>
>> The alternative is to try and help work on better speech synthesisers 
>> and bring something better forward and make that free software.
>
>
>
> </div>




More information about the Speakup mailing list