speakup and programming code

speakup and programming code

[Charles Hallenbeck]

Recent posts by Thomas Ward and others have prompted this
suggestion for a speakup feature that might greatly facilitate
things for people who deal with program code such as C or C++,
and might make exception dictionaries less critical to implement.

Many times the mixture of capitalization within an alphabetic
string is unusual in program code, and of course it is important.
I first ran across this many years ago with the name of the
popular data base package "d base ii". Now what you should have
heard inside the quotes is what everyone says when they pronounce
it, but "d base" is actually written "dBASE", and I defy you to
hear that correctly without spelling it out character by
character.

Here is my suggestion: When speakup is sending a series of
letters to the synth and notices that (1) the current char is
upper case, and (2) the last char sent was lower case, then (3)
before sending the current char it should send whatever is needed
to break the current string into two parts. Maybe that would be a
CR, or a space, or some unspoken control char, or whatever. The
result would be "d base" instead of "dBASE", and C programmers
will recognize immediately that there will be zillions of similar
funny case mixtures that will be spoken more correctly if the
transition from lower to upper case within a string is broken up
with a neutral unspoken element that serves only to cause the
synth to pronounce what it has already received and treat the
following as a new word.

This would perhaps be an inexpensive speakup modification that
would dramatically improve its performance for some of us.

Chuck


*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*
Visit me at http://www.mhonline.net/~chuckh
The Moon is Waning Crescent (1% of Full)

Re: speakup and programming code

[Saqib Shaikh]

I agree that this is a useful feature as i use it within emacspeak and jfw
extensively.  saqib
----- Original Message -----
From: "Charles Hallenbeck" <chuckh@mhonline.net>
To: "Speakup Distribution List" <speakup@speech.braille.uwo.ca>
Sent: Saturday, January 12, 2002 12:25 PM
Subject: speakup and programming code


> Recent posts by Thomas Ward and others have prompted this
> suggestion for a speakup feature that might greatly facilitate
> things for people who deal with program code such as C or C++,
> and might make exception dictionaries less critical to implement.
>
> Many times the mixture of capitalization within an alphabetic
> string is unusual in program code, and of course it is important.
> I first ran across this many years ago with the name of the
> popular data base package "d base ii". Now what you should have
> heard inside the quotes is what everyone says when they pronounce
> it, but "d base" is actually written "dBASE", and I defy you to
> hear that correctly without spelling it out character by
> character.
>
> Here is my suggestion: When speakup is sending a series of
> letters to the synth and notices that (1) the current char is
> upper case, and (2) the last char sent was lower case, then (3)
> before sending the current char it should send whatever is needed
> to break the current string into two parts. Maybe that would be a
> CR, or a space, or some unspoken control char, or whatever. The
> result would be "d base" instead of "dBASE", and C programmers
> will recognize immediately that there will be zillions of similar
> funny case mixtures that will be spoken more correctly if the
> transition from lower to upper case within a string is broken up
> with a neutral unspoken element that serves only to cause the
> synth to pronounce what it has already received and treat the
> following as a new word.
>
> This would perhaps be an inexpensive speakup modification that
> would dramatically improve its performance for some of us.
>
> Chuck
>
>
> *<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*
> Visit me at http://www.mhonline.net/~chuckh
> The Moon is Waning Crescent (1% of Full)
>
>
> _______________________________________________
> Speakup mailing list
> Speakup@braille.uwo.ca
> http://speech.braille.uwo.ca/mailman/listinfo/speakup

Re: speakup and programming code

[Gregory Nowak]

Grate sugestion, but it wouldn't work for everything. For example, something like "printf" begines with a lowercase p, and ends with a lowercase f, which would mean that it wouldn't get broken up if I understand your implementation of it.
Greg


On Sat, Jan 12, 2002 at 07:25:28AM -0500, Charles Hallenbeck wrote:
> Recent posts by Thomas Ward and others have prompted this
> suggestion for a speakup feature that might greatly facilitate
> things for people who deal with program code such as C or C++,
> and might make exception dictionaries less critical to implement.
> 
> Many times the mixture of capitalization within an alphabetic
> string is unusual in program code, and of course it is important.
> I first ran across this many years ago with the name of the
> popular data base package "d base ii". Now what you should have
> heard inside the quotes is what everyone says when they pronounce
> it, but "d base" is actually written "dBASE", and I defy you to
> hear that correctly without spelling it out character by
> character.
> 
> Here is my suggestion: When speakup is sending a series of
> letters to the synth and notices that (1) the current char is
> upper case, and (2) the last char sent was lower case, then (3)
> before sending the current char it should send whatever is needed
> to break the current string into two parts. Maybe that would be a
> CR, or a space, or some unspoken control char, or whatever. The
> result would be "d base" instead of "dBASE", and C programmers
> will recognize immediately that there will be zillions of similar
> funny case mixtures that will be spoken more correctly if the
> transition from lower to upper case within a string is broken up
> with a neutral unspoken element that serves only to cause the
> synth to pronounce what it has already received and treat the
> following as a new word.
> 
> This would perhaps be an inexpensive speakup modification that
> would dramatically improve its performance for some of us.
> 
> Chuck
> 
> 
> *<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*
> Visit me at http://www.mhonline.net/~chuckh
> The Moon is Waning Crescent (1% of Full)
> 
> 
> _______________________________________________
> Speakup mailing list
> Speakup@braille.uwo.ca
> http://speech.braille.uwo.ca/mailman/listinfo/speakup

Re: speakup and programming code

[Charles Hallenbeck]

Right. No solution is perfect, but this one gets a lot of
mileage.

On Sat, 12 Jan 2002, Gregory Nowak wrote:

> Grate sugestion, but it wouldn't work for everything. For example, something like "printf" begines with a lowercase p, and ends with a lowercase f, which would mean that it wouldn't get broken up if I understand your implementation of it.
> Greg
>
>
> On Sat, Jan 12, 2002 at 07:25:28AM -0500, Charles Hallenbeck wrote:
> > Recent posts by Thomas Ward and others have prompted this
> > suggestion for a speakup feature that might greatly facilitate
> > things for people who deal with program code such as C or C++,
> > and might make exception dictionaries less critical to implement.
> >
> > Many times the mixture of capitalization within an alphabetic
> > string is unusual in program code, and of course it is important.
> > I first ran across this many years ago with the name of the
> > popular data base package "d base ii". Now what you should have
> > heard inside the quotes is what everyone says when they pronounce
> > it, but "d base" is actually written "dBASE", and I defy you to
> > hear that correctly without spelling it out character by
> > character.
> >
> > Here is my suggestion: When speakup is sending a series of
> > letters to the synth and notices that (1) the current char is
> > upper case, and (2) the last char sent was lower case, then (3)
> > before sending the current char it should send whatever is needed
> > to break the current string into two parts. Maybe that would be a
> > CR, or a space, or some unspoken control char, or whatever. The
> > result would be "d base" instead of "dBASE", and C programmers
> > will recognize immediately that there will be zillions of similar
> > funny case mixtures that will be spoken more correctly if the
> > transition from lower to upper case within a string is broken up
> > with a neutral unspoken element that serves only to cause the
> > synth to pronounce what it has already received and treat the
> > following as a new word.
> >
> > This would perhaps be an inexpensive speakup modification that
> > would dramatically improve its performance for some of us.
> >
> > Chuck
> >
> >
> > *<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*
> > Visit me at http://www.mhonline.net/~chuckh
> > The Moon is Waning Crescent (1% of Full)
> >
> >
> > _______________________________________________
> > Speakup mailing list
> > Speakup@braille.uwo.ca
> > http://speech.braille.uwo.ca/mailman/listinfo/speakup
>
> _______________________________________________
> Speakup mailing list
> Speakup@braille.uwo.ca
> http://speech.braille.uwo.ca/mailman/listinfo/speakup
>

*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*
Visit me at http://www.mhonline.net/~chuckh
The Moon is Waning Crescent (1% of Full)

Re: speakup and programming code

[Buddy Brannan]

Yeah, and how 'bout adding a bit to split up ham radio calls so's they
get pronounced correctly? (There are enough hams on this list...lOL)
-- 
Buddy Brannan, KB5ELV    | I choose you to take up all of my time.
Email: davros@ycardz.com | I choose you because you're funny and kind
                         | I want easy people from now on.
                         | --the Nields

Re: speakup and programming code

[Thomas Ward]

Which brings us right back to the original problem. Since programming
languages are cap specific you would have to write printf as printf and cout
as cout.
I think that implamenting an exception dictionary, and put it in speakup
wouldn't be all that hard. However, I'm not familiar enough with Speakup to
do it myself.

----- Original Message -----
From: Charles Hallenbeck <chuckh@mhonline.net>
To: <speakup@braille.uwo.ca>
Sent: Saturday, January 12, 2002 11:23 AM
Subject: Re: speakup and programming code


> Right. No solution is perfect, but this one gets a lot of
> mileage.
>
> On Sat, 12 Jan 2002, Gregory Nowak wrote:
>
> > Grate sugestion, but it wouldn't work for everything. For example,
something like "printf" begines with a lowercase p, and ends with a
lowercase f, which would mean that it wouldn't get broken up if I understand
your implementation of it.
> > Greg
> >
> >
> > On Sat, Jan 12, 2002 at 07:25:28AM -0500, Charles Hallenbeck wrote:
> > > Recent posts by Thomas Ward and others have prompted this
> > > suggestion for a speakup feature that might greatly facilitate
> > > things for people who deal with program code such as C or C++,
> > > and might make exception dictionaries less critical to implement.
> > >
> > > Many times the mixture of capitalization within an alphabetic
> > > string is unusual in program code, and of course it is important.
> > > I first ran across this many years ago with the name of the
> > > popular data base package "d base ii". Now what you should have
> > > heard inside the quotes is what everyone says when they pronounce
> > > it, but "d base" is actually written "dBASE", and I defy you to
> > > hear that correctly without spelling it out character by
> > > character.
> > >
> > > Here is my suggestion: When speakup is sending a series of
> > > letters to the synth and notices that (1) the current char is
> > > upper case, and (2) the last char sent was lower case, then (3)
> > > before sending the current char it should send whatever is needed
> > > to break the current string into two parts. Maybe that would be a
> > > CR, or a space, or some unspoken control char, or whatever. The
> > > result would be "d base" instead of "dBASE", and C programmers
> > > will recognize immediately that there will be zillions of similar
> > > funny case mixtures that will be spoken more correctly if the
> > > transition from lower to upper case within a string is broken up
> > > with a neutral unspoken element that serves only to cause the
> > > synth to pronounce what it has already received and treat the
> > > following as a new word.
> > >
> > > This would perhaps be an inexpensive speakup modification that
> > > would dramatically improve its performance for some of us.
> > >
> > > Chuck
> > >
> > >
> > > *<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*
> > > Visit me at http://www.mhonline.net/~chuckh
> > > The Moon is Waning Crescent (1% of Full)
> > >
> > >
> > > _______________________________________________
> > > Speakup mailing list
> > > Speakup@braille.uwo.ca
> > > http://speech.braille.uwo.ca/mailman/listinfo/speakup
> >
> > _______________________________________________
> > Speakup mailing list
> > Speakup@braille.uwo.ca
> > http://speech.braille.uwo.ca/mailman/listinfo/speakup
> >
>
> *<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*
> Visit me at http://www.mhonline.net/~chuckh
> The Moon is Waning Crescent (1% of Full)
>
>
> _______________________________________________
> Speakup mailing list
> Speakup@braille.uwo.ca
> http://speech.braille.uwo.ca/mailman/listinfo/speakup
>

Re: speakup and programming code

[Steve Holmes]

Hey, I like that "mixed case" feature as well. This concept of
capitalizing the first letter of each implied word in program variables is
heavily stressed in coding standards for Visual Basic.  I rather like the
idea, even as a coding standard for other languages.  If a screen reader
can logically break these up into words and speak them separately, then it
is even that much better.

On Sat, 12 Jan 2002, Charles Hallenbeck wrote:

> Recent posts by Thomas Ward and others have prompted this
> suggestion for a speakup feature that might greatly facilitate
> things for people who deal with program code such as C or C++,
> and might make exception dictionaries less critical to implement.
>
> Many times the mixture of capitalization within an alphabetic
> string is unusual in program code, and of course it is important.
> I first ran across this many years ago with the name of the
> popular data base package "d base ii". Now what you should have
> heard inside the quotes is what everyone says when they pronounce
> it, but "d base" is actually written "dBASE", and I defy you to
> hear that correctly without spelling it out character by
> character.
>
> Here is my suggestion: When speakup is sending a series of
> letters to the synth and notices that (1) the current char is
> upper case, and (2) the last char sent was lower case, then (3)
> before sending the current char it should send whatever is needed
> to break the current string into two parts. Maybe that would be a
> CR, or a space, or some unspoken control char, or whatever. The
> result would be "d base" instead of "dBASE", and C programmers
> will recognize immediately that there will be zillions of similar
> funny case mixtures that will be spoken more correctly if the
> transition from lower to upper case within a string is broken up
> with a neutral unspoken element that serves only to cause the
> synth to pronounce what it has already received and treat the
> following as a new word.
>
> This would perhaps be an inexpensive speakup modification that
> would dramatically improve its performance for some of us.
>
> Chuck
>
>
> *<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*<<<=-=>>>*
> Visit me at http://www.mhonline.net/~chuckh
> The Moon is Waning Crescent (1% of Full)
>
>
> _______________________________________________
> Speakup mailing list
> Speakup@braille.uwo.ca
> http://speech.braille.uwo.ca/mailman/listinfo/speakup
>