Hi,

I've got a large sound file which is a dump from a voicemail box which
i'd like to split up into individual files, the only marker that I have
is the voicemail box system saying "Next Message". Unfortunatly there
isn't any DTMF tones in to separate on.

I've had a pretty good search around to find some software that will do
something like this:

Sample the "Next Message" portion of the file and then get some magic
piece of software to go though the file and identify it by the wave
form and mark out all the regions allowing me to then split out the
individual messages that I have.

My question is does this piece of software exist, and if so where can I
buy it?

Thanks in advance

wrote:
I've got a large sound file which is a dump from a voicemail box which
i'd like to split up into individual files, the only marker that I have
is the voicemail box system saying "Next Message". Unfortunatly there
isn't any DTMF tones in to separate on.

I've had a pretty good search around to find some software that will do
something like this:

Sample the "Next Message" portion of the file and then get some magic
piece of software to go though the file and identify it by the wave
form and mark out all the regions allowing me to then split out the
individual messages that I have.

My question is does this piece of software exist, and if so where can I
buy it?

If you know something about audio and you know how to program in
LISP, you could probably write a Nyquist plugin for Audacity to
do this in about 1 hour.

Alternately, if you know a little about sound and have a library
to read and write WAV files (which there are lots of), you could
probably write a standalone program to do it pretty easily.

All you've got to do is find approximate matches to the "next message"
sample, which you can do by subtracting samples and seeing if the
result is within some tolerance of zero. This ignores issues like
sample rate being off a bit, but for a small 1 or 2 second, it
probably doesn't matter.

- Logan