We've gotten an interesting request from one of our customers and I
figured I'd put it to the list to see if anyone has any idea how to
go about this.
The short version: They want callers to be able to say "skip" and
skip to the next of a series of sound files, but, if anything other
than "skip" is detected, continue (or resume playback). The problem
with this seems to be that if an utterance is detected, the speech
rec system stops sound playback, and advances in the dialplan. There
doesn't appear to be any way to resume playback of the file if
unrecognized or invalid speech is detected.
Has anyone encountered a need like this before? Does anyone have an
idea how to go about it? The C API looks like it might be promising,
but my C-fu is pretty weak.