Fix speed-perturbatin for swbd + execuation permissions for some fish…

…er_swbd scripts

egs/fisher_swbd/s5/local/chain/run_tdnn_lstm_1b.sh

@@ -28,8 +28,8 @@
 # Final valid prob (xent)       -0.9393   -0.9837
 
 #./steps/info/chain_dir_info.pl exp/chain/tdnn_lstm_1b_sp
-#exp/chain/tdnn_lstm_1b_sp: num-iters=1909 nj=3..16 num-params=39.7M dim=40+100->6149 combine=-0.087->-0.086 (over 5) 
-#xent:train/valid[1270,1908,final]=(-1.37,-1.02,-1.01/-1.31,-1.00,-0.984) 
+#exp/chain/tdnn_lstm_1b_sp: num-iters=1909 nj=3..16 num-params=39.7M dim=40+100->6149 combine=-0.087->-0.086 (over 5)
+#xent:train/valid[1270,1908,final]=(-1.37,-1.02,-1.01/-1.31,-1.00,-0.984)
 #logprob:train/valid[1270,1908,final]=(-0.108,-0.088,-0.087/-0.103,-0.091,-0.088)
 
 
@@ -108,7 +108,7 @@ dir=${dir}$suffix
 build_tree_train_set=train_nodup
 train_set=train_nodup_sp
 build_tree_ali_dir=exp/tri5a_ali
-treedir=exp/chain/tri6_tree
+treedir=exp/chain/tri5_tree
 lang=data/lang_chain
 
 # if we are using the speed-perturbed data we need to generate

egs/swbd/s5c/local/nnet3/run_ivector_common.sh

@@ -4,49 +4,38 @@
 set -e
 stage=1
 train_stage=-10
-generate_alignments=true # false if doing ctc training
+generate_alignments=true
 speed_perturb=true
 
 . ./path.sh
 . ./utils/parse_options.sh
 
-mkdir -p nnet3
-# perturbed data preparation
+mkdir -p exp/nnet3
 train_set=train_nodup
 
 if [ -e data/rt03 ]; then maybe_rt03=rt03; else maybe_rt03= ; fi
 
-if [ "$speed_perturb" == "true" ]; then
+if $speed_perturb; then
   if [ $stage -le 1 ]; then
-    #Although the nnet will be trained by high resolution data, we still have to perturbe the normal data to get the alignment
+    # Although the nnet will be trained by high resolution data, we still have to perturb the normal data to get the alignments
     # _sp stands for speed-perturbed
-
-    for datadir in train_nodup; do
-      utils/perturb_data_dir_speed.sh 0.9 data/${datadir} data/temp1
-      utils/perturb_data_dir_speed.sh 1.1 data/${datadir} data/temp2
-      utils/combine_data.sh data/${datadir}_tmp data/temp1 data/temp2
-      utils/validate_data_dir.sh --no-feats data/${datadir}_tmp
-      rm -r data/temp1 data/temp2
-
-      mfccdir=mfcc_perturbed
-      steps/make_mfcc.sh --cmd "$train_cmd" --nj 50 \
-        data/${datadir}_tmp exp/make_mfcc/${datadir}_tmp $mfccdir || exit 1;
-      steps/compute_cmvn_stats.sh data/${datadir}_tmp exp/make_mfcc/${datadir}_tmp $mfccdir || exit 1;
-      utils/fix_data_dir.sh data/${datadir}_tmp
-
-      utils/copy_data_dir.sh --spk-prefix sp1.0- --utt-prefix sp1.0- data/${datadir} data/temp0
-      utils/combine_data.sh data/${datadir}_sp data/${datadir}_tmp data/temp0
-      utils/fix_data_dir.sh data/${datadir}_sp
-      rm -r data/temp0 data/${datadir}_tmp
-    done
+    echo "$0: preparing directory for speed-perturbed data"
+    utils/data/perturb_data_dir_speed_3way.sh data/${train_set} data/${train_set}_sp
+
+    echo "$0: creating MFCC features for low-resolution speed-perturbed data"
+    mfccdir=mfcc_perturbed
+    steps/make_mfcc.sh --cmd "$train_cmd" --nj 50 \
+                       data/${train_set}_sp exp/make_mfcc/${train_set}_sp $mfccdir
+    steps/compute_cmvn_stats.sh data/${train_set}_sp exp/make_mfcc/${train_set}_sp $mfccdir
+    utils/fix_data_dir.sh data/${train_set}_sp
   fi
 
-  if [ $stage -le 2 ] && [ "$generate_alignments" == "true" ]; then
-    #obtain the alignment of the perturbed data
+  if [ $stage -le 2 ] && $generate_alignments; then
+    # obtain the alignment of the perturbed data
     steps/align_fmllr.sh --nj 100 --cmd "$train_cmd" \
-      data/train_nodup_sp data/lang exp/tri4 exp/tri4_ali_nodup_sp || exit 1
+      data/${train_set}_sp data/lang exp/tri4 exp/tri4_ali_nodup_sp
   fi
-  train_set=train_nodup_sp
+  train_set=${train_set}_sp
 fi
 
 if [ $stage -le 3 ]; then

egs/wsj/s5/utils/data/perturb_data_dir_speed_3way.sh

@@ -38,9 +38,17 @@ utils/data/get_utt2dur.sh ${srcdir}
 
 utils/data/perturb_data_dir_speed.sh 0.9 ${srcdir} ${destdir}_speed0.9 || exit 1
 utils/data/perturb_data_dir_speed.sh 1.1 ${srcdir} ${destdir}_speed1.1 || exit 1
-utils/data/combine_data.sh $destdir ${srcdir} ${destdir}_speed0.9 ${destdir}_speed1.1 || exit 1
 
-rm -r ${destdir}_speed0.9 ${destdir}_speed1.1
+utils/copy_data_dir.sh --spk-prefix sp1.0- --utt-prefix sp1.0- ${srcdir} data/temp0
+if [ ! -f $srcdir/utt2uniq ]; then
+  cat $srcdir/utt2spk | awk  '{printf("sp1.0-%s %s\n", $1, $1);}' > data/temp0/utt2uniq
+else
+  cat $srcdir/utt2uniq | awk '{printf("sp1.0-%s %s\n", $1, $2);}' > data/temp0/utt2uniq
+fi
+
+utils/data/combine_data.sh $destdir data/temp0 ${destdir}_speed0.9 ${destdir}_speed1.1 || exit 1
+
+rm -r ${destdir}_speed0.9 ${destdir}_speed1.1 data/temp0
 
 echo "$0: generated 3-way speed-perturbed version of data in $srcdir, in $destdir"
 utils/validate_data_dir.sh --no-feats --no-text $destdir