Speech Segmentation - Townshend Computer Tools DAT-Link User Manual

6.15 Speech Segmentation

Custom versions of the
narecord
that described in Section 5.14 for the
6.15 Speech Segmentation
DAT-Link
When the is used to record or transfer speech speech signals, it is often
desirable to segment the speech into words or segments. A front-end program for
the program called
narecord
As the speech is acquired by the
of each block of samples is also computed and passed to the
These energy computations allow
speech segment begins and ends. Then, along with a le containing the acquired
data, a le containing pointers to each segment is written by
of the segmentation le is described in Appendix B.2.
The program accepts the same
narecseg
accepted by . In addition, the
narecord
ments should be recorded before stopping. By default, one segment is recorded.
Note that the audio data is always stored using the
are not currently supported by
program can be created in a manner similar to
program.
naplay
provides this function.
narecseg
DAT-Link
's signal processor, the average energy
to heuristically determine where each
narecseg
, , ,
-u -a -i
can be used to specify how many seg-
-n
raw
.
narecseg
program.
narecseg
. The format
narecseg
, and options that are
-o -s
le format. Other formats
69