志浩
2023-08-02 a0ffc9ba6b86780aa68ceca8b41bce464cf2a066
TOLD/SOND: modify data preprocessing
2个文件已修改
1个文件已添加
36 ■■■■ 已修改文件
egs/callhome/diarization/sond/conf/fbank.conf 4 ●●●● 补丁 | 查看 | 原始文档 | blame | 历史
egs/callhome/diarization/sond/finetune.sh 29 ●●●● 补丁 | 查看 | 原始文档 | blame | 历史
egs/callhome/diarization/sond/run.sh 3 ●●●● 补丁 | 查看 | 原始文档 | blame | 历史
egs/callhome/diarization/sond/conf/fbank.conf
New file
@@ -0,0 +1,4 @@
--sample-frequency=8000
--num-mel-bins=80
--frame-length=25
--snip-edges=false
egs/callhome/diarization/sond/finetune.sh
@@ -92,8 +92,21 @@
# Prepare datasets
if [ ${stage} -le 0 ] && [ ${stop_stage} -ge 0 ]; then
  echp "Stage 0: Prepare callhome data."
  echo "Stage 0: Prepare callhome data."
  local/make_callhome.sh ${callhome_root} ${datadir}/
  # split ref.rttm
  for dset in callhome1 callhome2; do
    rm -rf data/${dset}/ref.rttm
    for name in `awk '{print $1}' data/${dset}/wav.scp`; do
      grep ${name} data/callhome/fullref.rttm >> data/${dset}/ref.rttm;
    done
    # filter out records which don't have rttm labels.
    awk '{print $2}' data/${dset}/ref.rttm | sort | uniq > data/${dset}/uttid
    mv data/${dset}/wav.scp data/${dset}/wav.scp.bak
    awk '{if (NR==FNR){a[$1]=1}else{if (a[$1]==1){print $0}}}' data/${dset}/uttid data/${dset}/wav.scp.bak > data/${dset}/wav.scp
  done
fi
if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then
@@ -123,10 +136,10 @@
    mkdir -p ${dumpdir}/${dset}/nonoverlap_0s
    python -Wignore script/extract_nonoverlap_segments.py \
      ${datadir}/${dset}/wav.scp ${datadir}/${dset}/ref.rttm ${dumpdir}/${dset}/nonoverlap_0s \
      --min_dur 0 --max_spk_num 8 --sr ${sr} --no_pbar --nj ${nj}
      --min_dur 0.1 --max_spk_num 8 --sr ${sr} --no_pbar --nj ${nj}
    mkdir -p ${datadir}/${dset}/nonoverlap_0s
    find `pwd`/${dumpdir}/${dset}/nonoverlap_0s | sort | awk -F'[/.]' '{print $(NF-1),$0}' > ${datadir}/${dset}/nonoverlap_0s/wav.scp
    find ${dumpdir}/${dset}/nonoverlap_0s/ -iname "*.wav" | sort | awk -F'[/.]' '{print $(NF-1),$0}' > ${datadir}/${dset}/nonoverlap_0s/wav.scp
    awk -F'[/.]' '{print $(NF-1),$(NF-2)}' ${datadir}/${dset}/nonoverlap_0s/wav.scp > ${datadir}/${dset}/nonoverlap_0s/utt2spk
    echo "Done."
  done
@@ -134,13 +147,17 @@
if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
  echo "Stage 3: Generate fbank features"
  home_path=`pwd`
  home_path=$(pwd)
  cd ${kaldi_root}/egs/callhome_diarization/v2 || exit
  . ./cmd.sh
  export train_cmd="run.pl"
  export cmd="run.pl"
  . ./path.sh
  cd $home_path || exit
  ln -s ${kaldi_root}/egs/callhome_diarization/v2/steps ./
  for dset in callhome1 callhome2; do
    mv ${datadir}/${dset}/segments ${datadir}/${dset}/segs
    steps/make_fbank.sh --write-utt2num-frames true --fbank-config conf/fbank.conf --nj ${nj} --cmd "$train_cmd" \
        ${datadir}/${dset} ${expdir}/make_fbank/${dset} ${dumpdir}/${dset}/fbank
    utils/fix_data_dir.sh ${datadir}/${dset}
@@ -151,8 +168,8 @@
        ${datadir}/${dset} ${expdir}/make_fbank/${dset} ${dumpdir}/${dset}/fbank
    utils/fix_data_dir.sh ${datadir}/${dset}
  done
  rm -f steps
  cd ${home_path} || exit
fi
if [ ${stage} -le 4 ] && [ ${stop_stage} -ge 4 ]; then
egs/callhome/diarization/sond/run.sh
@@ -48,6 +48,7 @@
# dataset related
data_root=
callhome_root=path/to/NIST/LDC2001S97
# experiment configuration
lang=en
@@ -124,7 +125,7 @@
  utils/fix_data_dir.sh ${datadir}/swbd_sre
  # 3. Prepare the Callhome portion of NIST SRE 2000.
  local/make_callhome.sh /nfs/wangjiaming.wjm/speech-data/NIST/LDC2001S97 ${datadir}/
  local/make_callhome.sh ${callhome_root} ${datadir}/
fi