Hi Adam, On Wed, Sep 30, 2009 at 07:20:32PM -0500, Adam Myrow wrote: > As to making it a choice between pitch shifting or a message, I don't see > how that's any different from what we are doing now, since, if the message > is in the i18n files, the user could customize it to anything they wanted. > They could even change it to the strings that are sent now if they felt so > inclined. Doing that would not be consistent with the way the other drivers work, so it would require reworking all of the speech drivers. > Let me ask another question here. What exactly happens internally in > Speakup when you press the various keys to adjust pitch or speed? For > example, insert+3 and insert+4 adjust the pitch. What I hear with the > Dectalk in these cases is a message like "pitch 104." The pitch is set to > that new value before the message is spoken. I ask, because it seems to > me that some of the same code could be used to do the pitch shifting > instead of sending a hard-coded string. That code sets the pitch exactly to a value that we keep track of in the speakup program. We know the range pitches have, and we currently pick a default of of 100 out of the air and start the dectalk at that default. Those keys just add or subtract one to that value and tell the synthesizer to reset it's pitch to the new value. Pitch shifting would be something like giving the synthesizer a command to add or subtract a value from whatever the current pitch setting is. This is what the dectalks do not support. So, right now, what the caps start/stop commands do is set the pitch to specific values, but this causes issues as well, because, if you change the voice to something other than Paul, when you change the pitch, you mess up the voice. That is what I am trying to avoid. What do you think? William
