public inbox for speakup@linux-speakup.org
 help / color / mirror / Atom feed
* Re: speakup and cepstral? (fwd)
@  Cheryl Homiak
   ` Cheryl Homiak
   ` Hynek Hanke
  0 siblings, 2 replies; 29+ messages in thread
From: Cheryl Homiak @  UTC (permalink / raw)
  To: speakup



-- 
Cheryl
"Where your treasure is, there will your heart be also."


---------- Forwarded message ----------
Date: Tue, 4 May 2004 12:43:35 -0500 (CDT)
From: Cheryl Homiak <chomiak@charter.net>
To: speakup <speakup@braille.uwo.ca>
Subject: Re: speakup and cepstral?

Ok, I have it primitively working.
Below is what I've done to the cpstrl-generic.conf file. I decided to
eliminate things I didn't know how to handle yet, and add them back as I
figure it out. This works from the command prompt, does Pine, reads things
in more (haven't tried less), and cursor-tracking and numpad functions
seem to be functional. Works in pico or nano but I lose speech completely
when I load emacs and it doesn't come back until i kill speechd_up and
speech-dispatcher and start them again. Echoes keystrokes but does not say
word after space but does say full completion when tab-completion is used.
At one time speech also stopped when I did control-x in pine and went into
the spell checker, but in another instance it didn't.
If I try to insert
-r $RATE -S $PITCH
and change the forceInteger lines from 1 to 0, I don't get any speech.
I realize that what I've done and what I've left out still leaves a lot to
e desired, but it still is progress. I haven't turned debugging on yet to
see what else I can find out; that might help me track down what's going
on when I load emacs.
I've only been using this a few minutes, so in addition to not handling
pitch and rate and the things I've mentioned above, there are probably all
sorts of problems. But the voice is really clear.
Further help and instruction will be eagerly accepted.
I've cut off the top of the .conf file for the sake of length.

GenericExecuteSynth \
"/usr/local/bin/theta -N $VOICE \"$DATA\"

# GenericStripPunctChars is a list (enclosed in doublequotes) of
# all the characters that should be replaced by whitespaces in
# order not to be badly handled by the output module or misinterpreted
# by shell.
# We need to strip `[' and `]', as these are DECTalk's
# command characters.
GenericStripPunctChars  "[]"

# AddVoice specifies which $VOICE string should be assigned to
# each language and symbolic voice name. All the voices you want
# to use must be specified here.
# NOTE:
# There is a multilingual version of DECTalk software, however I
# do not have it. Thus, only the US English voices are defined here.
# If you know about the other languages, please let us know on
# <speechd@freebsoft.org>

AddVoice	"en"	"MALE1"		"Frank"
# AddVoice	"en"	"MALE2"		"harry"
# AddVoice	"en"	"MALE3"		"dennis"
# AddVoice	"en"	"FEMALE1" 	"betty"
# AddVoice	"en"	"FEMALE2"	"ursula"
# AddVoice	"en"	"FEMALE3"	"wendy"
# AddVoice	"en"	"CHILD_MALE"	"kit"

# These parameters set _rate_ and _pitch_ conversion. This is
# part of the core of the definition of this generic output
# module for this concrete synthesizer, it's not intended to
# be modified by common users.
# The resulting rate (or pitch) has the form:
# 	(speechd_rate * GenericRateMultiply) + GenericRateAdd
# while speechd_rate is a value between -100 (lowest) and +100 (highest)
# You have to define some meaningful conversion for each synthesizer
# NOTE:
# Because DECTalk cannot accept float values, we must force them to be
# integers.

GenericRateForceInteger 1
GenericPitchForceInteger 1

GenericRateAdd 338
GenericPitchAdd	225

# (These values are multiplied by 100, because DotConf currently
# doesn't support floats. So you can write 0.85 as 85 and so on.)

GenericRateMultiply 262
GenericPitchMultiply 175

# Debug turns debugging on or off
# Debug 1

# DebugFile specifies the file where the debugging information
# should be stored (note that the log is overwritten each time
# the module starts)
# DebugFile "/tmp/debug-dtk-generic"

-- 
Cheryl
"Where your treasure is, there will your heart be also."




^ permalink raw reply	[flat|nested] 29+ messages in thread
* speakup and cepstral?
@  Cheryl Homiak
  0 siblings, 0 replies; 29+ messages in thread
From: Cheryl Homiak @  UTC (permalink / raw)
  To: speakup

Now that the list is back up, I'm going to re-send my posts on this topic.
I've gotten it at least basically talking, but right now am having trouble
with how to implement rate and pitch variables. as you'll see from one of
my re--sends, cepstral does support floats like epos and unlike dectalk.
there are other problems, but this is what I'm trying to solve right
now--pitch and rate.
Will just resend posts with possibly a little editing.



-- 
Cheryl
"Where your treasure is, there will your heart be also."



^ permalink raw reply	[flat|nested] 29+ messages in thread
* speakup and cepstral?
@  Cheryl Homiak
  0 siblings, 0 replies; 29+ messages in thread
From: Cheryl Homiak @  UTC (permalink / raw)
  To: speakup

I just bought one of the voices for cepstral and am wondering if it meets
the qualifications for use with speech-dispatcher and speakup. I have no
experience with this kind of thing and don't know how to proceed, so
please don't laugh at me for what I did as a preliminary experiment. I
copied dtk-generic.conf to a file called cpstrl-generic.conf and changed
/usr/local/bin/say to /usr/local/bin/theta, but I didn't know what else
would have to be done to map the variables so just for starters I did
nothing to them. Then I added the cpstrl-generic to my AddModules with the
sd_generic module and the appropriate .conf file and made this the
default.
it actually came up speaking, but it wants to interpret and repeat the
variables on the line with it; for instance it doesn't say $Voice; it says
something about Paul instead, and gives the pitch and rate numbers. but it
also does repeat one's input and the screen's output, so I think there
might be help. I do realize that what I did was a very elementary
experiment; i was gratified that it gave me any speech output at all.
I'm not trying to push this off on anybody else to work on; in fact, I'd
like to try it, but I might be in over my head. would somebody be willing
to give me an idea what steps I should take in trying to get this to
work? Now, if anybody knows that for
some reason this can't be done, go ahead and let me know so I don't waste
my time and drive everybody on the list crazy.
I'm using the "Frank" voice and it sounds quite good. I may buy other
voices eventually but thought one voice would be a good start.
By the way, cepstral does a nice job with wave files.




-- 
Cheryl
"Where your treasure is, there will your heart be also."



^ permalink raw reply	[flat|nested] 29+ messages in thread
end of thread, other threads:[~ UTC | newest]

Thread overview: 29+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
 speakup and cepstral? (fwd) Cheryl Homiak
 ` Cheryl Homiak
 ` Hynek Hanke
   ` speakup and cepstral? Steve Holmes
     ` Cheryl Homiak
       ` Cheryl Homiak
       ` Cheryl Homiak
         ` Steve Holmes
           ` Cheryl Homiak
             ` speakup and Flite 16 Bit Voices Richard Wells
               ` Steve Holmes
                 ` Richard Wells
             ` speakup and cepstral? Steve Holmes
     ` nick G
       ` Steve Holmes
     ` Flite and different voices (was: Re: speakup and cepstral?) Hynek Hanke
       ` Flite and different voices Steve Holmes
         ` Cheryl Homiak
         ` Steve Holmes
           ` Jacob Schmude
           ` Hynek Hanke
         ` Steve Holmes
           ` Hynek Hanke
             ` Steve Holmes
               ` nick G
     ` speakup and cepstral? Garrett Klein
       ` Chuck Hallenbeck
  -- strict thread matches above, loose matches on Subject: below --
 Cheryl Homiak
 Cheryl Homiak

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).