def run(self, inputfilename, outputphonfile=None, outputtokensfile=None):
""" Run the Phonetization process from a tokenized input.TO DO.
Parameters:
- inputfilename is the input file name
- outputphonfile is the output file name of the phonetization
- outputtokensfile is the output file name of the tokenization
Return: None
Exceptions: IOError
"""
if not outputphonfile:
outputphonfile = inputfilename.replace(".TextGrid", "-phon.TextGrid")
if not outputtokensfile:
outputtokensfile = inputfilename.replace(".TextGrid", "-tokens.TextGrid")
# Program names and data dirs
esppasdir = os.path.join(SPPAS, "scripts")
esppasdir = os.path.join(esppasdir, "esppas")
formate = os.path.join(esppasdir, "formatphonFRTOE.awk")
rustine = os.path.join(esppasdir, "rustinePhonFRTOE.sh")
# ALGORITHM:
# - EXTRACT EACH INTERVAL
# - CALL LIA_PHON ON THIS INTERVAL
# - GET RESULT AND PUT IN AN OUTPUT TIER
# Write the appropriate tier in a LIA-style text file
# it also will convert input from UTF8 to iso8859-1
try:
inputok = genName().get_name() + ".liatxt"
trs = Transcription()
trs.Append(inputfilename.get_tier(1))
annotationdata.io.write(inputok, trs)
except Exception:
raise BaseException("Phon::esppasphon.py. Processing of phonetization failed.\n")
# Execute LIA_Phon
command = "cat "
command += inputok
command += " | $LIA_PHON_REP/script/lia_text2phon > "
command += tmpoutput
# ESPPAS patches
command = "cat "
command += tmpoutput
command += " | gawk -f "
command += formate
command += " -l -t -m 250 -s '##' | "
command += rustine
command += " > "
command += outputphonfile
ret = self.run_command(command)
# self.run_command( "rm "+tmpoutput )
if not os.path.isfile(outputphonfile):
raise BaseException("Phon::esppasphon.py. Post-processing of TOE-Phonetization failed.\n")
def old_run(self, inputfilename, outputphonfile=None, outputtokensfile=None):
""" Run the Phonetization process from a transcription.
For details, see:
B. Bigi, P. Péri, R. Bertrand (2012)
Orthographic Transcription: which Enrichment is required for phonetization?
Language Resources and Evaluation Conference, Istanbul (Turkey),
pages 1756-1763, ISBN 978-2-9517408-7-7.
INPUT: an enriched orthographic transcription
BRIEF: Get the TO, create a tree, call LIA_Tagg, update the tree,
call LIA_Phon, update the tree, then get the phonetization.
Parameters:
- inputfilename is the input file name
- outputphonfile is the output file name of the phonetization
- outputtokensfile is the output file name of the tokenization
Return: None
Exceptions: IOError
"""
if not outputphonfile:
outputphonfile = inputfilename.replace(".TextGrid", "-phon.TextGrid")
if not outputtokensfile:
outputtokensfile = inputfilename.replace(".TextGrid", "-tokens.TextGrid")
# Program names and data dirs
esppasdir = os.path.join(SPPAS, "scripts")
esppasdir = os.path.join(esppasdir, "esppas")
jar = os.path.join(esppasdir, "phonetizationFRTOE.jar")
etc = os.path.join(esppasdir, "etc")
etc = os.path.join(etc, " ")
formate = os.path.join(esppasdir, "formatphonFRTOE.awk")
rustine = os.path.join(esppasdir, "rustinePhonFRTOE.sh")
# First command to execute
command = "java -Dfile.encoding=iso8859-1 -jar "
command += jar
# with its parameters:
command += " -s " + etc
command += " -i " + inputfilename
command += " -t 1 "
tmpoutput = genName().get_name() + ".TextGrid"
command += " -o " + tmpoutput
command += " -w " + outputtokensfile
command += " -v 0 "
ret = self.run_command(command)
if not os.path.isfile(tmpoutput):
raise BaseException("Phon::esppasphon.py. TOE-Phonetization failed.\n")
# ESPPAS patches
command = "cat "
command += tmpoutput
command += " | gawk -f "
command += formate
command += " -l -t -m 250 -s '##' | "
command += rustine
command += " > "
command += outputphonfile
ret = self.run_command(command)
# self.run_command( "rm "+tmpoutput )
if not os.path.isfile(outputphonfile):
raise BaseException("Phon::esppasphon.py. Post-processing of TOE-Phonetization failed.\n")