Skip to content
Snippets Groups Projects

Compare revisions

Changes are shown as if the source revision was being merged into the target revision. Learn more about comparing revisions.

Source

Select target project
No results found

Target

Select target project
  • deniz.cingoez/robocup-speechrec
  • t.dilger/robocup-speechrec
  • msterz/robocup-speechrec
  • patric.steckstor/robocup-speechrec
  • jennifer.lee/robocup-speechrec
  • jharin/robocup-speechrec
  • f.thiemer/robocup-speechrec
  • augustin.harter/robocup-speechrec
  • jleichert/robocup-speechrec
9 results
Show changes
Hide whitespace changes
Inline Side-by-side
Showing
with 0 additions and 870 deletions
isr-grammar/english/phrecog/isr/phrec.cl deleted 120000 → 0
View file @ b5cf02de
../wsj1/wsj1.cl
\ No newline at end of file
isr-grammar/english/phrecog/isr/phrec.lex deleted 100644 → 0
View file @ b5cf02de
/3/
/3r/
/@/
/A/
/D/
/E/
/I/
/N/
/O/
/Oi/
/S/
/T/
/U/
/V/
/Z/
/aI/
/aU/
/b/
/d/
/dZ/
/eI/
/f/
/g/
/h/
/i/
/j/
/k/
/l/
/m/
/n/
/oU/
/p/
/r/
/s/
/t/
/tS/
/u/
/v/
/w/
/z/
<sil>
isr-grammar/english/phrecog/isr/phrec.swu deleted 100644 → 0
View file @ b5cf02de
Source diff could not be displayed: it is too large. Options to address this: view the blob.
isr-grammar/english/phrecog/isr/phrec.tree deleted 100644 → 0
View file @ b5cf02de
# - -
#./3/ /3/ /3/
#./3r/ /3r/ /3r/
#./@/ /@/ /@/
#./A/ /A/ /A/
#./D/ /D/ /D/
#./E/ /E/ /E/
#./I/ /I/ /I/
#./N/ /N/ /N/
#./O/ /O/ /O/
#./Oi/ /Oi/ /Oi/
#./S/ /S/ /S/
#./T/ /T/ /T/
#./U/ /U/ /U/
#./V/ /V/ /V/
#./Z/ /Z/ /Z/
#./aI/ /aI/ /aI/
#./aU/ /aU/ /aU/
#./b/ /b/ /b/
#./d/ /d/ /d/
#./dZ/ /dZ/ /dZ/
#./eI/ /eI/ /eI/
#./f/ /f/ /f/
#./g/ /g/ /g/
#./h/ /h/ /h/
#./i/ /i/ /i/
#./j/ /j/ /j/
#./k/ /k/ /k/
#./l/ /l/ /l/
#./m/ /m/ /m/
#./n/ /n/ /n/
#./oU/ /oU/ /oU/
#./p/ /p/ /p/
#./r/ /r/ /r/
#./s/ /s/ /s/
#./t/ /t/ /t/
#./tS/ /tS/ /tS/
#./u/ /u/ /u/
#./v/ /v/ /v/
#./w/ /w/ /w/
#./z/ /z/ /z/
#.[-] [-] <sil>
#.[--] [--] <sil>
#.[---] [---] <sil>
isr-grammar/english/phrecog/isr/rec.sh deleted 100755 → 0
View file @ b5cf02de
arecord -r16000 -c1 -traw -fS16_LE
isr-grammar/english/phrecog/wsj1 deleted 120000 → 0
View file @ b5cf02de
/vol/esmeralda/share/models/wsj1/
\ No newline at end of file
isr-grammar/german/Kunsthalle2011/Kunsthalle.ch deleted 100644 → 0
View file @ b5cf02de
0 4 4.75
2.66537 0.460283 0.722534 0.881632 0.771319 0.647087 0.547872 0.471176 0.691693 0.834649 0.694927 0.725007 0.7158 0.72206 0.759329 0.839217 1.01634 0.996053 0.997086 1.02442 0.990176 1.00615 1.17781 1.25187 1.37796 1.54784 1.39192 1.39529 1.43478 1.31847 1.21826 1.76494 1.13124 0.951468
0 0.35
0.1 0.6
0.2 0.95
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.25 1
0.2 1
0.3 1
0.3 0.95
0.2 1
0.25 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.25 1
0.2 1
0.2 1
0.25 1
0.25 1
0.25 1
0.2 1
0.25 1
0.25 1
0.25 1
0.2 1
0.2 1
0.2 1
0.2 1
0.25 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.25 1
0.25 1
0.3 1
0.25 0.95
0.2 1
0.25 1
0.2 1
0.25 1
0.3 1
0.15 0.95
0.25 1
0.25 1
0.25 1
0.25 1
0.2 1
0.25 1
0.25 1
0.25 1
0.2 1
0.2 1
0.3 1
0.2 1
0.2 1
0.25 1
0.2 1
0.25 1
0.2 1
0.2 1
0.2 1
0.25 1
0.25 1
0.25 1
0.25 1
0.2 1
0.2 1
0.2 1
0.25 1
0.25 1
0.25 1
0.3 1
0.25 1
0.2 1
0.2 1
0.25 1
0.25 1
0.2 1
0.2 1
0.25 1
0.2 1
0.2 1
0.25 1
0.2 1
0.25 1
0.2 1
0.25 1
0.2 1
0.25 1
0.2 1
0.2 1
0.25 1
0.2 0.95
0.25 1
0.2 1
0.15 1
0.15 1
0.2 1
0.25 1
0.25 1
0.25 1
0.2 1
0.25 1
0.2 1
0.2 1
0.25 1
0.15 1
0.15 1
0.25 1
0.2 1
0.2 1
0.3 1
0.2 1
0.25 1
0.2 1
0.2 1
0.25 1
0.25 1
0.2 1
0.2 1
0.2 1
0.25 1
0.2 1
0.2 1
0.25 1
0.2 1
0.25 1
0.25 1
0.2 1
0.25 1
0.2 1
0.2 1
0.25 1
0.2 1
0.2 1
0.25 1
0.25 1
0.25 1
0.2 1
0.2 1
0.25 1
0.2 1
0.25 1
0.2 1
0.25 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.15 1
0.2 1
0.15 1
0.2 1
0.2 1
0.2 1
0.25 1
0.2 1
0.2 1
0.2 1
0.15 1
0.2 1
0.15 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.2 1
0.25 1
0.3 1
0.3 1
0.55 0.55
0.85 0.85
0.7 0.7
0.35 1
0.25 1
0.25 1
0.9 0.9
0.2 1
0.15 1
0.25 1
0.3 1
0.2 1
0.2 1
0.2 1
0.2 0.95
0.15 0.95
0.15 0.95
0.15 0.95
0.15 0.95
0.15 0.95
0.15 0.95
0.15 0.9
0.1 0.9
0.15 0.9
0.15 0.9
0.15 0.95
0.15 0.95
0.15 0.9
0.15 0.95
0.15 0.9
0.15 0.9
0.15 0.9
0.15 0.95
0.15 0.9
0.15 0.95
0.15 0.9
0.15 0.9
0.15 0.9
0.1 0.9
0.1 0.9
0.15 0.9
0.1 0.85
0.1 0.85
0.1 0.8
0.1 0.85
0.1 0.85
0.1 0.85
0.1 0.8
0.1 0.8
0.1 0.8
0.1 0.75
0.1 0.7
0.05 0.65
0.1 0.65
0.1 0.65
0.05 0.65
0.05 0.6
0.05 0.5
0.05 0.45
0.05 0.45
0.05 0.4
0.05 0.35
0.05 0.35
0.05 0.3
0.05 0.3
0.05 0.25
0 0.25
0 0.25
0 0.2
0 0.25
isr-grammar/german/Kunsthalle2011/Kunsthalle.cl deleted 120000 → 0
View file @ b5cf02de
/vol/esmeralda/share/models/vm/vm.cl
\ No newline at end of file
isr-grammar/german/Kunsthalle2011/Kunsthalle.grm deleted 100644 → 0
View file @ b5cf02de
$$S: $answer | $Greeting ;
$answer: $not_ok | $ok ;
$ok: $ok_prefix | $ok_prefix $ok_modifier | $ok_modifier ;
$ok_prefix: ja | ja ja | jo | japp | joa ;
$ok_modifier: gerne | bitte | klar | meinetwegen | von mir aus | mach das | mach doch | los | bitte sch"on | das d"urfen sie | bitte das w"urde mich freuen | wenn du das kannst | sehr gern | bitte gerne ;
$not_ok: $not_ok_prefix | $not_ok_prefix $not_ok_modifier | $not_ok_modifier ;
$not_ok_prefix: nein | nee | n"o ;
$not_ok_modifier: danke | bitte nicht | gerade nicht | jetzt nicht | danke ich will nicht | wir waren schon hier | hast du schon | nicht nochmal | das hast du doch gerade schon gemacht | hast du doch schon ;
$Greeting: $greeting_prefix | $greeting_prefix $robot_name ;
$greeting_prefix: hallo | guten tag | hey | hi ;
$robot_name: roboter | du | sie | nao ;
isr-grammar/german/Kunsthalle2011/Kunsthalle.phono deleted 100644 → 0
View file @ b5cf02de
Aufladestation aUf|la:|d@|Sta||tsjo:n
Nachrichten na:x|rIC|tn
Satz zats
Bruder brU||d6
Lied li:d
aufzuzeichnen aUf|tsu|tsaIC|n
Sklaverei skla!|fE6|aI
Telefonanrufe te|l@|fo:n|an|ru:|f@
Telefonanruf te|l@|fo:n|an|ru:|f
aufzunehmen ?aUf|tsu:|ne:|m@n
geh ge:
zeichnen tsaIC|n@n
aufzunehmen ?aUf|tsu:|ne:|m@n
Hi haI
Nao na:|o:
Jo jo:
Japp ja!p
Joa joa
isr-grammar/german/Kunsthalle2011/MakeVars.models deleted 100644 → 0
View file @ b5cf02de
# MakeVars for biron german
# grammar name
NAME=Kunsthalle
# acoustic base model
BASE=vm
# will not be transcribed
LEXADDON=Noises.addon.lex
# phonetic transcriptions
PHONOS=Total.phono Kunsthalle.phono
# language for installation target
LANGUAGE=de
isr-grammar/german/Kunsthalle2011/Makefile.models deleted 120000 → 0
View file @ b5cf02de
../../Makefile.models
\ No newline at end of file
isr-grammar/german/Kunsthalle2011/Noises.addon.lex deleted 100644 → 0
View file @ b5cf02de
isr-grammar/german/Kunsthalle2011/Total.phono deleted 120000 → 0
View file @ b5cf02de
/vol/esmeralda/share/models/vm/Total.phono
\ No newline at end of file
isr-grammar/german/Kunsthalle2011/build.sh deleted 100755 → 0
View file @ b5cf02de
#!/bin/bash
make -f Makefile.models remove
make -f Makefile.models
isr-grammar/german/Kunsthalle2011/isr.sh deleted 100755 → 0
View file @ b5cf02de
#CFGDIR=.
#name=zahlen
#CFGDIR=/vol/esmeralda/share/configurations/SpeechRec/english/NaoVFOAQuiz
#name=NaoVFOAQuiz
CFGDIR=/vol/esmeralda/share/configurations/SpeechRec/german/Kunsthalle2011
name=Kunsthalle
####
# Configurations:
NGRAM="bi-gram"
#NGRAM="tri-gram"
LM_SWITCH=0
GRM_SWITCH=1
###
if [ $LM_SWITCH -eq 1 ]; then
CODE="lm-$NGRAM"
fi
if [ $GRM_SWITCH -eq 1 ]; then
CODE="$CODE-grm"
fi
#in=telefon.test.$USER.ssg.rcs
#out=telefon.test.$CODE
# Options for isr
#OPTS="-b70 -l4 -B30 -T5 -o -"
#OPTS="-b70 -l4 -B30 -o xcf:isr"
OPTS="-b70 -l4 -B30 -o rsb:isr -m rsb"
# -b : beam width
# -l : linguistic matching factor
# -B : language model beam with
# -T : enable tracing of currently best result every <n> frames
# -o : output hypotheses to <name>
# specifies a file that holds sub-word unit definitions
SWU=$CFGDIR/$name.swu
# specifies a file containing the recognition lexicon
LEX=$CFGDIR/$name.lex
# specifies a file with the recognition model tree
TREE=$CFGDIR/$name.tree
# [LM]
# language model (grammar)
#GRM_OPTS="-w 10 -f 5 -i 5 -I -x 0.3 -p ALL"
GRM_OPTS="-w 10 -f 4 -i 6 -I -x 0.3 -p ALL"
# -w : word penalty
# -f : penalty for abortions.
# -i : penalty for initialisations.
# -I : makes initialisation penalty independent of context.
# -x : weighting factor for structure expansions [0.0-1.0].
if [ $GRM_SWITCH -eq 1 ]; then
GRM="-g $CFGDIR/$name.grm"
# -g : specifies a file holding the grammar definition
fi
# [LM] cont.
LM_OPTS=
LM=-
if [ $LM_SWITCH -eq 1 ]; then
# specifies a file holding the language model definition
LM=$CFGDIR/$name.$NGRAM.lm
fi
# [CL]
# specifies a file with a code-library definition
CL=$CFGDIR/$name.cl
# [AA]
AA=-
# [FE]
# feature extraction
#FE_OPTS=""
# -p <param> configure feature extraction using parameters <param>
FE_OPTS="-p $CFGDIR/$name.ch"
#FE="mfcc 1.4"
FE="mfcc 1.6"
# specifies the type of feature extraction to use.
# specifies the version of the feature extraction to be used.
# [SR]
# -s30 -u35
SR_OPTS="-B 32 -r4.5 -s45 -u46"
# -B <n> set speech segment recording buffer size to <n> segments
# -r <x> set reference energy for calculating db to <x>
# -s <x> set start-utterance energy threshold to <x> db
# -u <x> set within-utterance energy threshold to <x> db
# -x send signal status information via xcf/rsb/...
#SR="$1"
SR="-"
# specifies the type of the signal source to be used
# Direct from normal/default sound card:
command="arecord -t raw -fS16_LE -r16000 -c1 | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# From external Edirol USB soundcard (two persons as one person ;) ):
#command="arecord -r 16000 -c 2 -f S16_LE -t wav -D pulse | sox --buffer 1024 -t wav - -t raw -c 1 - remix 2 | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# nc -l -u -p 12345 | sox -t wav -c 4 - -t wav -s -2 -L -c 2 - remix 2v2 1v2 noisered naonoise.profile 0.25
# From jack:
#command="/homes/dklotz/nao/speech/jack-stdout jack.play-10453:out_3 | sox -t raw -r 48000 -c 1 -s -2 - -t raw -c 1 -s -2 - rate -m 16k | /vol/nao/releases/cuttingedge/bin/isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# Using the RSB audio receiver:
#command="rsbaudioreceiver | /usr/bin/sox -t raw -r 16000 -c 1 -s -2 - -t raw -c 1 -s -2 - | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# Direct network transmission using netcat:
#command="nc -l -u -p 12345 | sox -t wav -c 4 - -t raw -s -2 -L -c 1 - remix 3v2 noisered /homes/dklotz/nao/speech/noise/naonoise_mono_front.profile 0.15 rate -m 16k | /vol/nao/bin/isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
echo $command
eval $command
isr-grammar/german/Kunsthalle2011/isr1.sh deleted 100755 → 0
View file @ b5cf02de
#CFGDIR=.
#name=zahlen
#CFGDIR=/vol/esmeralda/share/configurations/SpeechRec/english/NaoVFOAQuiz
#name=NaoVFOAQuiz
CFGDIR=/vol/esmeralda/share/configurations/SpeechRec/german/Kunsthalle2011
name=Kunsthalle
####
# Configurations:
NGRAM="bi-gram"
#NGRAM="tri-gram"
LM_SWITCH=0
GRM_SWITCH=1
###
if [ $LM_SWITCH -eq 1 ]; then
CODE="lm-$NGRAM"
fi
if [ $GRM_SWITCH -eq 1 ]; then
CODE="$CODE-grm"
fi
#in=telefon.test.$USER.ssg.rcs
#out=telefon.test.$CODE
# Options for isr
#OPTS="-b70 -l4 -B30 -T5 -o -"
#OPTS="-b70 -l4 -B30 -o xcf:isr"
OPTS="-b70 -l4 -B30 -o rsb:isr -m rsb"
# -b : beam width
# -l : linguistic matching factor
# -B : language model beam with
# -T : enable tracing of currently best result every <n> frames
# -o : output hypotheses to <name>
# specifies a file that holds sub-word unit definitions
SWU=$CFGDIR/$name.swu
# specifies a file containing the recognition lexicon
LEX=$CFGDIR/$name.lex
# specifies a file with the recognition model tree
TREE=$CFGDIR/$name.tree
# [LM]
# language model (grammar)
#GRM_OPTS="-w 10 -f 5 -i 5 -I -x 0.3 -p ALL"
GRM_OPTS="-w 10 -f 4 -i 6 -I -x 0.3 -p ALL"
# -w : word penalty
# -f : penalty for abortions.
# -i : penalty for initialisations.
# -I : makes initialisation penalty independent of context.
# -x : weighting factor for structure expansions [0.0-1.0].
if [ $GRM_SWITCH -eq 1 ]; then
GRM="-g $CFGDIR/$name.grm"
# -g : specifies a file holding the grammar definition
fi
# [LM] cont.
LM_OPTS=
LM=-
if [ $LM_SWITCH -eq 1 ]; then
# specifies a file holding the language model definition
LM=$CFGDIR/$name.$NGRAM.lm
fi
# [CL]
# specifies a file with a code-library definition
CL=$CFGDIR/$name.cl
# [AA]
AA=-
# [FE]
# feature extraction
#FE_OPTS=""
# -p <param> configure feature extraction using parameters <param>
FE_OPTS="-p $CFGDIR/$name.ch"
#FE="mfcc 1.4"
FE="mfcc 1.6"
# specifies the type of feature extraction to use.
# specifies the version of the feature extraction to be used.
# [SR]
# -s30 -u35
SR_OPTS="-B 32 -r4.5 -s45 -u46"
# -B <n> set speech segment recording buffer size to <n> segments
# -r <x> set reference energy for calculating db to <x>
# -s <x> set start-utterance energy threshold to <x> db
# -u <x> set within-utterance energy threshold to <x> db
# -x send signal status information via xcf/rsb/...
#SR="$1"
SR="-"
# specifies the type of the signal source to be used
# Direct from normal/default sound card:
#command="arecord -t raw -fS16_LE -r16000 -c1 | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# From external Edirol USB soundcard (just the first channel):
command="arecord -r 16000 -c 2 -f S16_LE -t wav -D pulse | sox --buffer 1024 -t wav - -t raw -c 1 - remix 1 | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# nc -l -u -p 12345 | sox -t wav -c 4 - -t wav -s -2 -L -c 2 - remix 2v2 1v2 noisered naonoise.profile 0.25
# From jack:
#command="/homes/dklotz/nao/speech/jack-stdout jack.play-10453:out_3 | sox -t raw -r 48000 -c 1 -s -2 - -t raw -c 1 -s -2 - rate -m 16k | /vol/nao/releases/cuttingedge/bin/isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# Using the RSB audio receiver:
#command="rsbaudioreceiver | /usr/bin/sox -t raw -r 16000 -c 1 -s -2 - -t raw -c 1 -s -2 - | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# Direct network transmission using netcat:
#command="nc -l -u -p 12345 | sox -t wav -c 4 - -t raw -s -2 -L -c 1 - remix 3v2 noisered /homes/dklotz/nao/speech/noise/naonoise_mono_front.profile 0.15 rate -m 16k | /vol/nao/bin/isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
echo $command
eval $command
isr-grammar/german/Kunsthalle2011/isr2.sh deleted 100755 → 0
View file @ b5cf02de
#CFGDIR=.
#name=zahlen
#CFGDIR=/vol/esmeralda/share/configurations/SpeechRec/english/NaoVFOAQuiz
#name=NaoVFOAQuiz
CFGDIR=/vol/esmeralda/share/configurations/SpeechRec/german/Kunsthalle2011
name=Kunsthalle
####
# Configurations:
NGRAM="bi-gram"
#NGRAM="tri-gram"
LM_SWITCH=0
GRM_SWITCH=1
###
if [ $LM_SWITCH -eq 1 ]; then
CODE="lm-$NGRAM"
fi
if [ $GRM_SWITCH -eq 1 ]; then
CODE="$CODE-grm"
fi
#in=telefon.test.$USER.ssg.rcs
#out=telefon.test.$CODE
# Options for isr
#OPTS="-b70 -l4 -B30 -T5 -o -"
#OPTS="-b70 -l4 -B30 -o xcf:isr"
OPTS="-b70 -l4 -B30 -o rsb:isr -m rsb"
# -b : beam width
# -l : linguistic matching factor
# -B : language model beam with
# -T : enable tracing of currently best result every <n> frames
# -o : output hypotheses to <name>
# specifies a file that holds sub-word unit definitions
SWU=$CFGDIR/$name.swu
# specifies a file containing the recognition lexicon
LEX=$CFGDIR/$name.lex
# specifies a file with the recognition model tree
TREE=$CFGDIR/$name.tree
# [LM]
# language model (grammar)
#GRM_OPTS="-w 10 -f 5 -i 5 -I -x 0.3 -p ALL"
GRM_OPTS="-w 10 -f 4 -i 6 -I -x 0.3 -p ALL"
# -w : word penalty
# -f : penalty for abortions.
# -i : penalty for initialisations.
# -I : makes initialisation penalty independent of context.
# -x : weighting factor for structure expansions [0.0-1.0].
if [ $GRM_SWITCH -eq 1 ]; then
GRM="-g $CFGDIR/$name.grm"
# -g : specifies a file holding the grammar definition
fi
# [LM] cont.
LM_OPTS=
LM=-
if [ $LM_SWITCH -eq 1 ]; then
# specifies a file holding the language model definition
LM=$CFGDIR/$name.$NGRAM.lm
fi
# [CL]
# specifies a file with a code-library definition
CL=$CFGDIR/$name.cl
# [AA]
AA=-
# [FE]
# feature extraction
#FE_OPTS=""
# -p <param> configure feature extraction using parameters <param>
FE_OPTS="-p $CFGDIR/$name.ch"
#FE="mfcc 1.4"
FE="mfcc 1.6"
# specifies the type of feature extraction to use.
# specifies the version of the feature extraction to be used.
# [SR]
# -s30 -u35
SR_OPTS="-B 32 -r4.5 -s45 -u46"
# -B <n> set speech segment recording buffer size to <n> segments
# -r <x> set reference energy for calculating db to <x>
# -s <x> set start-utterance energy threshold to <x> db
# -u <x> set within-utterance energy threshold to <x> db
# -x send signal status information via xcf/rsb/...
#SR="$1"
SR="-"
# specifies the type of the signal source to be used
# Direct from normal/default sound card:
#command="arecord -t raw -fS16_LE -r16000 -c1 | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# From external Edirol USB soundcard (just the second channel):
command="arecord -r 16000 -c 2 -f S16_LE -t wav -D pulse | sox --buffer 1024 -t wav - -t raw -c 1 - remix 2 | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# nc -l -u -p 12345 | sox -t wav -c 4 - -t wav -s -2 -L -c 2 - remix 2v2 1v2 noisered naonoise.profile 0.25
# From jack:
#command="/homes/dklotz/nao/speech/jack-stdout jack.play-10453:out_3 | sox -t raw -r 48000 -c 1 -s -2 - -t raw -c 1 -s -2 - rate -m 16k | /vol/nao/releases/cuttingedge/bin/isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# Using the RSB audio receiver:
#command="rsbaudioreceiver | /usr/bin/sox -t raw -r 16000 -c 1 -s -2 - -t raw -c 1 -s -2 - | isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
# Direct network transmission using netcat:
#command="nc -l -u -p 12345 | sox -t wav -c 4 - -t raw -s -2 -L -c 1 - remix 3v2 noisered /homes/dklotz/nao/speech/noise/naonoise_mono_front.profile 0.15 rate -m 16k | /vol/nao/bin/isr $OPTS $SWU $LEX $TREE $GRM_OPTS $GRM $LM_OPTS $LM $CL $AA $FE_OPTS $FE $SR_OPTS $SR"
echo $command
eval $command