Rather than translating dtk sequences to sable,
the cleaner solution is to write two sable specific .el
modules,
see the emacspeak sources for how I did this for outloud,
and clone that.
>>>>> "Mario" == Mario Lang <mlang@xxxxxxxxxxx> writes:
Mario> Hello. As some off you probably know, I am in
Mario> the process of writing a Emacspeak Server for
Mario> Festival. I am currently multithreading the
Mario> whole thing to be able to eliminate commands
Mario> which wouldn't get spoken (sch as fast scrolling
Mario> with many q d s q d s q d s series)... When this
Mario> is finished (I hope soon, its quite a brainer for
Mario> me), the next step will be the voice-lock mode.
Mario> I plan to convert the strings emacspeak sends to
Mario> the speech server to a SABLE-mark-up text for
Mario> Festival.
Mario> First of all, is this the right way? Or does
Mario> anyone think that implementing the SABLE commands
Mario> in a festival-voices.el would be better?
Mario> Ok, if q1 is no, than I seek for help: I'd like
Mario> to talk with someone how to map the DECtalk
Mario> commandset with all his parameters to apropriate
Mario> SABLE commands and their parameters. This can
Mario> only be well done from the ground up. I am
Mario> absolutly not familiar with DECtalk and dont know
Mario> what those many parameters (the numbers)
Mario> mean. Didnt find any good specs anyway..
Mario> Something to think about: SABLE supports:
Mario> <SPEAKER></SPEAKER>, <RATE></RATE>,
Mario> <SAYAS></SAYAS>, <AUDIO SRC/> Thats the main
Mario> commands to map to. The speech rate has to be in
Mario> the range of 1 to 99%. Speakers are speaker
Mario> names. SAYAS can be used for pronounciation and
Mario> so on. AUDIO SRC can be used to include Wavforms
Mario> directly into the spoken text.
Mario> Just got another question: Does anyone know how
Mario> to reconfigure Emacspeak so that it dosent use
Mario> the play program for Auditory Icons. It should
Mario> use the a and p commands and send Auditory Icons
Mario> to the Speech server. This would eliminates the
Mario> "two-soundcards" problem.
Mario> And the last question mainly to raman: Are you
Mario> planning to implement a pronounce-word command
Mario> for the speech servers? When I tell Emacspeak to
Mario> read the characters of a word it would send: l
Mario> {x} l {y} l {z} and so on. My speech server
Mario> would send the following to festival: (tts
Mario> "<SABLE><SAYAS MODE="literal">x</SAYAS> </SABLE>"
Mario> 'sable) (tts "<SABLE><SAYAS
Mario> MODE="literal">y</SAYAS> </SABLE>" 'sable) (tts
Mario> "<SABLE><SAYAS MODE="literal">z</SAYAS> </SABLE>"
Mario> 'sable)
Mario> But if Emacspeak would send a special command for
Mario> speaking literals of a word, I could convert to:
Mario> (tts "<SABLE><SAYAS MODE="literal">xyz</SAYAS>
Mario> </SABLE>" 'sable)
Mario> which would recude overhead alot.
Mario> Regards, Mario Lang <lang@xxxxxxxxxxx-graz.ac.at>
Mario> -----------------------------------------------------------------------------
Mario> To unsubscribe or change your address send mail
Mario> to "emacspeak-request@xxxxxxxxxxx" with a
Mario> subject of "unsubscribe" or "help"
--
Best Regards,
--raman
------------------------------------------------------------
IBM Research: Human Language Technologies
Phone: 1 (408) 927 2608
Fax: 1 (408) 927 3102
Email: tvraman@xxxxxxxxxxx
WWW: http://cs.cornell.edu/home/raman
PGP: http://cs.cornell.edu/home/raman/raman.asc
Snail: IBM Almaden Research Center,
650 Harry Road
San Jose 95120