[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: tuning via voice

Here is the patch file for outloud-voices.el.  

I have some comments below...

"T. V. Raman" <raman@users.sf.net> writes:

> See the blog article I posted on tuning TTS a couple of weeks ago;
> there is little written on this topic.

So I've found.  I guess I was wondering if there was something more
objective.  I know that aural qualities are mostly subjective, but I was
hoping there was a "best practices" sort of thing.

> There is Janet Cahn's MS thesis from about 13+ years ago, and there is
> my PhD thesis -- the latter is online.

I am looking at it right now.

My original comments are still true.  In addition, I have made no
changes to any settings other than pitch range.

>     Robert> I've not yet made any changes to anything over a5.
>     Robert> 
>     Robert> I still think the setting for a1, voice-bolden, is too
>     Robert> low.  If someone creates a www page that has a passage
>     Robert> done in bold, it can be annoying.  I have left it this low
>     Robert> since the frequency of this is low and to bring it any
>     Robert> higher will make the setting for a2 worthless.

In using these settings, I have almost gotten used to the voice for
bolden.  I do, however, think it might still be too low.




Robert D. Crawford                                      rdc1x@comcast.net

Your reasoning powers are good, and you are a fairly good planner.

Emacspeak Files | Subscribe | Unsubscribe | Search