Re: status of speakup support for espeak

* Re: status of speakup support for espeak
@  Hynek Hanke
   ` eSpeak and Festival Jonathan Duddington
                   ` (2 more replies)
  0 siblings, 3 replies; 16+ messages in thread
From: Hynek Hanke @  UTC (permalink / raw)
  To: speakup

Steve Holmes wrote:
> Yeah but Espeak is much smaller and frankly better than festival.
>   

Hello Steve and others,

me personally and some other people here can't really
understand how can someone consider the espeak voices
be of better quality than the festival voices. To me it
seems the difference in quality is just about incomparable.

At least, it seems to be a matter of user preference.
But given the difference is really quite big, I think
it might be something more as well. Perhaps the ones
who tested Festival didn't test the better quality
voices or their setup just included 8-bit voices
or perhaps there was some other problem. This
would be a no surprise because given that Festival
is a more complex solution and flexible solution,
it is necessarily a bit harder to setup and there
are more things that one can get wrong.

Another think that might be in play here is that
I expect that many blind users are used to the
simple hardware synthesizers, from which many
used to offer quite an artificial speech. Perhaps
these users got used to this sound so much that
they now actually consider a more natural synthesized
voice as of lower quality?

It would be great if somebody who thinks that Festival
is actually worse than eSpeak in quality of speech
could try to elaborate more about the reasons. We might
then try to discover if these reasons are real or if
they are of user preference or if it is even possible
to fix them with a slight fix in the configuration.
I think such an effort would be very useful. To discover
the possible traps as well as to get more light on this
discussion which, I admit, seems always very strange
to me. All help appreciated.

Now do not get me wrong that I think eSpeak is a bad
program, because that is totally not the case. eSpeak
offers (in my opinion) a lower speech quality as a trade-off
for very high speed, small size, disponibility with voices for a great
amount of languages and easy installation/very little configuration
necessary. The level of support for accessibility is very good
in both of them (Festival through festival-freebsoft-utils, espeak
natively).

This is why eSpeak is the current default in Speech Dispatcher
because it is initially easier to get running and it covers a great
span of languages. The documentation however strongly suggest
users whose language is supported by Festival to try it as their
primary syntesizer for a better voice quality.

With regards,
Hynek Hanke

^ permalink raw reply	[flat|nested] 16+ messages in thread