At 8:34 PM -0700 9/16/02, you wrote: >My project has text highlighting on screen following what's being >spoken in audio. I'm using cue points in wav files and Lingo to trap >for the right moments to do what I need to do. > >The problem is that a few of the audio files won't send their cue >points (or Lingo doesn't receive them) as they happen. > >Say there's a file that's 3 seconds long with five words. I put cue >points in the wav file before each word is spoken and another after >the last word is spoken. That makes six cue points. > >This allows me to highlight the word to be spoken as it is about to happen. > >Here's an example of what happens in the order I receive the messages: > >cue1 - first cue point, right before sound is spoken >the - the word is spoken >cue2 - next cue received >rain - spoken >cue3 - another cue received on time >in - word spoken >spain - another word spoken, next cue not received >fell - still no cue4 for the previous word >cue4 - finally the cue point from a couple words back >cue5 - and now the rest of the cue points >cue6 - cues 4,5 and 6 all come in immediate succession > >I've tried entering the cue points in three different editors, >changing their positions, massaging them in whatever way I can come >up with. There is one thing that tends to help. If I add silence at >the end of the file, the situation improves. But it might require >3/4 of a second beyond the .15-.25 that was there to begin with. >This causes some inconvenience as the wait gets in the way of other >events that follow. > >All the files with cue points are structured the same way, the voice >is the same on all of them. I've tried this in various forms, >playing the sounds as external sound files (my preferred method), as >cast members and using on cuePassed to get the cue points as they go >by even when playing the sounds directly from the cast. Within the >project, I'm using sound(channel).mostRecentCuePoint to get the data >from the wav files, but the problem is the same no matter what Lingo >I use or how I play the sounds. member(theMember).cuePointNames and >member().cuePointTimes show the proper info. > >Does anyone have any experience with such behavior? Any >consideration is much appreciated.
you wrote: "I'm using sound(channel).mostRecentCuePoint to get the data from the wav files," this can certainly cause the problem you have described. Checking the mostRecentCuePoint() periodically can often miss points when they occur close together. That method checks after the event has happened. It doesn't tell you when it occurred, only that it did, at some time in the past. As Irv has mentioned, if you put your response code in an 'on cuePassed handler', you should get responses as every cue occurs. hth -Buzz [To remove yourself from this list, or to change to digest mode, go to http://www.penworks.com/lingo-l.cgi To post messages to the list, email [EMAIL PROTECTED] (Problems, email [EMAIL PROTECTED]). Lingo-L is for learning and helping with programming Lingo. Thanks!]
