Skip to content

correct rttm script, add lab script #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jukaradayi wants to merge 1 commit into
base: master
Choose a base branch
from
Open

correct rttm script, add lab script #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
49 changes: 49 additions & 0 deletions run-lab.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,49 @@
#!/bin/bash
#
# Run script to generate HTK MFCC features, given a WAV audio file
# then given a speech/nonspeech file (extension .rttm), run DiarTK
# (also known as ib_diarization_toolkit) to produce RTTM clustered
# utterances and generated speaker IDs

# Assumes 10ms frame size in .scp file; to change, edit line in htkconfig:
# TARGETRATE = 100000.0


numargs=3
if [ $# -lt $numargs ]; then
echo "Usage: run-rttm.sh ipfile rttmfile outdir"
echo ""
echo "ipfile : Audio input file in WAV format, extension .wav"
echo "rttmfile: speech/nonspeech file, format:"
echo " Type file chan tbeg tdur ortho stype name conf Slat"
echo "produces output in folder outdir/"
exit
fi

filename=$(basename "$1")
basename="${filename%.*}"

workdir=$3

mkdir -p $workdir

featfile=$workdir/$basename.fea
scpfile=$workdir/$basename.scp

# first-first convert RTTM to DiarTK's version of a .scp file
# SCP format:
# <basename>_<start>_<end>=<filename>[start,end]
# RTTM format:
# Type file chan tbeg tdur ortho stype name conf Slat
# math: convert lab seconds to HTK (10ms default) frames = multiply by 100
grep " speech" $2 | awk -v base="$basename" -v feats="$featfile" '{begg=$1*100;endd=$2*100; print base "_" begg "_" endd "="feats "[" begg "," endd "]"}' > $scpfile

# first generate HTK features
HCopy -T 2 -C htkconfig $1 $featfile

# next run DiarTK
scripts/run.diarizeme.sh $featfile $scpfile $workdir $basename

# print results
cat $workdir/$basename.out

3 changes: 1 addition & 2 deletions run-rttm.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,9 +36,8 @@ scpfile=$workdir/$basename.scp
# RTTM format:
# Type file chan tbeg tdur ortho stype name conf Slat
# math: convert RTTM seconds to HTK (10ms default) frames = multiply by 100
grep LEXEME $2 | awk -v base="$basename" -v feats="$featfile" '{begg=$4*100;endd=($4+$5)*100; print base "_" begg "_" endd "="feats "[" begg "," endd "]"}' > $scpfile
grep SPEAKER $2 | awk -v base="$basename" -v feats="$featfile" '{begg=$4*100;endd=($4+$5)*100; print base "_" begg "_" endd "="feats "[" begg "," endd "]"}' > $scpfile

exit
# first generate HTK features
HCopy -T 2 -C htkconfig $1 $featfile

Expand Down