The software package SAMTRA normalises and verifies
transcriptions in SAMPA notation and computes phoneme and
diphoneme distribution statistics. The data flow between the
three modules of SAMTRA is via the intermediate files SAMTRA.NOR
and SAMTRA.STA (which are accessible for further developments).
The modules perform the following actions:
Module 1: Transcription verification
Module 2: Computation of phoneme and diphoneme
Module 3: Flexible printing of statistics
SAMTRA reads two input files, a SAMPA symbol file and a
SAMPA transcription file, and tries to write a normalised
transcription to the file SAMPTA.NOR.
If there are errors in the transcription file, SAMTRA
prints them out on the screen and to the file SAMTRA.NOR.
If the normalisation procedure succeeds, SAMTRA reads the
normalised transcription file SAMTRA.NOR and computes the
phoneme and diphoneme distributions, and writes them to
The text output of SAMTRA is dependent on the file
SAMTRA.SET (which overrides the default printout constants) and
command line switches.
Faculty of Linguistics and Literary Studies
D - 33501 Bielefeld