[scripts,egs] Replace LDA layer with delta and delta-delta features (#…

…3490)

WER is about the same but this is simpler to implement and more standard.

egs/mini_librispeech/s5/local/chain/run_tdnn.sh

@@ -1 +1 @@
-tuning/run_tdnn_1i.sh
+tuning/run_tdnn_1j.sh

egs/mini_librispeech/s5/local/chain/tuning/run_tdnn_1j.sh

@@ -0,0 +1,301 @@
+#!/bin/bash
+
+# 1j is as 1i but replaces the LDA layer at the input of the
+# network with delta and delta-delta features.
+# Below, 1j2 is just a different rerun of 1j with different
+# --affix option, to give some idea of the run-to-run variation.
+
+# local/chain/compare_wer.sh --online exp/chain/tdnn1i_sp exp/chain/tdnn1j_sp exp/chain/tdnn1j2_sp
+# System                tdnn1i_sp tdnn1j_sp tdnn1j2_sp
+#WER dev_clean_2 (tgsmall)      10.73     10.78     10.79
+#             [online:]         10.81     10.80     10.80
+#WER dev_clean_2 (tglarge)       7.54      7.43      7.54
+#             [online:]          7.59      7.37      7.55
+# Final train prob        -0.0621   -0.0620   -0.0615
+# Final valid prob        -0.0787   -0.0787   -0.0785
+# Final train prob (xent)   -1.4603   -1.4363   -1.4438
+# Final valid prob (xent)   -1.5847   -1.5629   -1.5712
+# Num-params                 5210944   5210944   5210944
+
+# steps/info/chain_dir_info.pl exp/chain/tdnn1j_sp
+# exp/chain/tdnn1j_sp: num-iters=34 nj=2..5 num-params=5.2M dim=40+100->2336 combine=-0.049->-0.046 (over 4) xent:train/valid[21,33,final]=(-1.40,-1.18,-1.17/-1.56,-1.38,-1.37) logprob:train/valid[21,33,final]=(-0.064,-0.052,-0.047/-0.089,-0.085,-0.078)
+
+# Set -e here so that we catch if any executable fails immediately
+set -euo pipefail
+
+# First the options that are passed through to run_ivector_common.sh
+# (some of which are also used in this script directly).
+stage=0
+decode_nj=10
+train_set=train_clean_5
+test_sets=dev_clean_2
+gmm=tri3b
+nnet3_affix=
+
+# The rest are configs specific to this script.  Most of the parameters
+# are just hardcoded at this level, in the commands below.
+affix=1j   # affix for the TDNN directory name
+tree_affix=
+train_stage=-10
+get_egs_stage=-10
+decode_iter=
+
+# training options
+# training chunk-options
+chunk_width=140,100,160
+common_egs_dir=
+xent_regularize=0.1
+
+# training options
+srand=0
+remove_egs=true
+reporting_email=
+
+#decode options
+test_online_decoding=true  # if true, it will run the last decoding stage.
+
+
+# End configuration section.
+echo "$0 $@"  # Print the command line for logging
+
+. ./cmd.sh
+. ./path.sh
+. ./utils/parse_options.sh
+
+if ! cuda-compiled; then
+  cat <<EOF && exit 1
+This script is intended to be used with GPUs but you have not compiled Kaldi with CUDA
+If you want to use GPUs (and have them), go to src/, and configure and make on a machine
+where "nvcc" is installed.
+EOF
+fi
+
+# The iVector-extraction and feature-dumping parts are the same as the standard
+# nnet3 setup, and you can skip them by setting "--stage 11" if you have already
+# run those things.
+local/nnet3/run_ivector_common.sh --stage $stage \
+                                  --train-set $train_set \
+                                  --gmm $gmm \
+                                  --nnet3-affix "$nnet3_affix" || exit 1;
+
+# Problem: We have removed the "train_" prefix of our training set in
+# the alignment directory names! Bad!
+gmm_dir=exp/$gmm
+ali_dir=exp/${gmm}_ali_${train_set}_sp
+tree_dir=exp/chain${nnet3_affix}/tree_sp${tree_affix:+_$tree_affix}
+lang=data/lang_chain
+lat_dir=exp/chain${nnet3_affix}/${gmm}_${train_set}_sp_lats
+dir=exp/chain${nnet3_affix}/tdnn${affix}_sp
+train_data_dir=data/${train_set}_sp_hires
+lores_train_data_dir=data/${train_set}_sp
+train_ivector_dir=exp/nnet3${nnet3_affix}/ivectors_${train_set}_sp_hires
+
+for f in $gmm_dir/final.mdl $train_data_dir/feats.scp $train_ivector_dir/ivector_online.scp \
+    $lores_train_data_dir/feats.scp $ali_dir/ali.1.gz; do
+  [ ! -f $f ] && echo "$0: expected file $f to exist" && exit 1
+done
+
+if [ $stage -le 10 ]; then
+  echo "$0: creating lang directory $lang with chain-type topology"
+  # Create a version of the lang/ directory that has one state per phone in the
+  # topo file. [note, it really has two states.. the first one is only repeated
+  # once, the second one has zero or more repeats.]
+  if [ -d $lang ]; then
+    if [ $lang/L.fst -nt data/lang/L.fst ]; then
+      echo "$0: $lang already exists, not overwriting it; continuing"
+    else
+      echo "$0: $lang already exists and seems to be older than data/lang..."
+      echo " ... not sure what to do.  Exiting."
+      exit 1;
+    fi
+  else
+    cp -r data/lang $lang
+    silphonelist=$(cat $lang/phones/silence.csl) || exit 1;
+    nonsilphonelist=$(cat $lang/phones/nonsilence.csl) || exit 1;
+    # Use our special topology... note that later on may have to tune this
+    # topology.
+    steps/nnet3/chain/gen_topo.py $nonsilphonelist $silphonelist >$lang/topo
+  fi
+fi
+
+if [ $stage -le 11 ]; then
+  # Get the alignments as lattices (gives the chain training more freedom).
+  # use the same num-jobs as the alignments
+  steps/align_fmllr_lats.sh --nj 75 --cmd "$train_cmd" ${lores_train_data_dir} \
+    data/lang $gmm_dir $lat_dir
+  rm $lat_dir/fsts.*.gz # save space
+fi
+
+if [ $stage -le 12 ]; then
+  # Build a tree using our new topology.  We know we have alignments for the
+  # speed-perturbed data (local/nnet3/run_ivector_common.sh made them), so use
+  # those.  The num-leaves is always somewhat less than the num-leaves from
+  # the GMM baseline.
+   if [ -f $tree_dir/final.mdl ]; then
+     echo "$0: $tree_dir/final.mdl already exists, refusing to overwrite it."
+     exit 1;
+  fi
+  steps/nnet3/chain/build_tree.sh \
+    --frame-subsampling-factor 3 \
+    --context-opts "--context-width=2 --central-position=1" \
+    --cmd "$train_cmd" 3500 ${lores_train_data_dir} \
+    $lang $ali_dir $tree_dir
+fi
+
+
+if [ $stage -le 13 ]; then
+  mkdir -p $dir
+  echo "$0: creating neural net configs using the xconfig parser";
+
+  num_targets=$(tree-info $tree_dir/tree |grep num-pdfs|awk '{print $2}')
+  learning_rate_factor=$(echo "print (0.5/$xent_regularize)" | python)
+
+  tdnn_opts="l2-regularize=0.03"
+  tdnnf_opts="l2-regularize=0.03 bypass-scale=0.66"
+  linear_opts="l2-regularize=0.03 orthonormal-constraint=-1.0"
+  prefinal_opts="l2-regularize=0.03"
+  output_opts="l2-regularize=0.015"
+
+  mkdir -p $dir/configs
+  cat <<EOF > $dir/configs/network.xconfig
+  input dim=100 name=ivector
+  input dim=40 name=input
+
+  # this takes the MFCCs and generates filterbank coefficients.  The MFCCs
+  # are more compressible so we prefer to dump the MFCCs to disk rather
+  # than filterbanks.
+  idct-layer name=idct input=input dim=40 cepstral-lifter=22 affine-transform-file=$dir/configs/idct.mat
+  batchnorm-component name=batchnorm0 input=idct
+  spec-augment-layer name=spec-augment freq-max-proportion=0.5 time-zeroed-proportion=0.2 time-mask-max-frames=20
+  
+  delta-layer name=delta input=spec-augment
+  no-op-component name=input2 input=Append(delta, Scale(0.4, ReplaceIndex(ivector, t, 0)))
+
+  # the first splicing is moved before the lda layer, so no splicing here
+  relu-batchnorm-layer name=tdnn1 $tdnn_opts dim=768 input=input2
+  tdnnf-layer name=tdnnf2 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=1
+  tdnnf-layer name=tdnnf3 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=1
+  tdnnf-layer name=tdnnf4 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=1
+  tdnnf-layer name=tdnnf5 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=0
+  tdnnf-layer name=tdnnf6 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=3
+  tdnnf-layer name=tdnnf7 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=3
+  tdnnf-layer name=tdnnf8 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=3
+  tdnnf-layer name=tdnnf9 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=3
+  tdnnf-layer name=tdnnf10 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=3
+  tdnnf-layer name=tdnnf11 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=3
+  tdnnf-layer name=tdnnf12 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=3
+  tdnnf-layer name=tdnnf13 $tdnnf_opts dim=768 bottleneck-dim=96 time-stride=3
+  linear-component name=prefinal-l dim=192 $linear_opts
+
+  ## adding the layers for chain branch
+  prefinal-layer name=prefinal-chain input=prefinal-l $prefinal_opts small-dim=192 big-dim=768
+  output-layer name=output include-log-softmax=false dim=$num_targets $output_opts
+
+  # adding the layers for xent branch
+  prefinal-layer name=prefinal-xent input=prefinal-l $prefinal_opts small-dim=192 big-dim=768
+  output-layer name=output-xent dim=$num_targets learning-rate-factor=$learning_rate_factor $output_opts
+EOF
+  steps/nnet3/xconfig_to_configs.py --xconfig-file $dir/configs/network.xconfig --config-dir $dir/configs/
+fi
+
+
+if [ $stage -le 14 ]; then
+  if [[ $(hostname -f) == *.clsp.jhu.edu ]] && [ ! -d $dir/egs/storage ]; then
+    utils/create_split_dir.pl \
+     /export/b0{3,4,5,6}/$USER/kaldi-data/egs/mini_librispeech-$(date +'%m_%d_%H_%M')/s5/$dir/egs/storage $dir/egs/storage
+  fi
+
+  steps/nnet3/chain/train.py --stage=$train_stage \
+    --cmd="$decode_cmd" \
+    --feat.online-ivector-dir=$train_ivector_dir \
+    --feat.cmvn-opts="--norm-means=false --norm-vars=false" \
+    --chain.xent-regularize $xent_regularize \
+    --chain.leaky-hmm-coefficient=0.1 \
+    --chain.l2-regularize=0.0 \
+    --chain.apply-deriv-weights=false \
+    --chain.lm-opts="--num-extra-lm-states=2000" \
+    --trainer.add-option="--optimization.memory-compression-level=2" \
+    --trainer.srand=$srand \
+    --trainer.max-param-change=2.0 \
+    --trainer.num-epochs=20 \
+    --trainer.frames-per-iter=3000000 \
+    --trainer.optimization.num-jobs-initial=2 \
+    --trainer.optimization.num-jobs-final=5 \
+    --trainer.optimization.initial-effective-lrate=0.002 \
+    --trainer.optimization.final-effective-lrate=0.0002 \
+    --trainer.num-chunk-per-minibatch=128,64 \
+    --egs.chunk-width=$chunk_width \
+    --egs.dir="$common_egs_dir" \
+    --egs.opts="--frames-overlap-per-eg 0" \
+    --cleanup.remove-egs=$remove_egs \
+    --use-gpu=true \
+    --reporting.email="$reporting_email" \
+    --feat-dir=$train_data_dir \
+    --tree-dir=$tree_dir \
+    --lat-dir=$lat_dir \
+    --dir=$dir  || exit 1;
+fi
+
+if [ $stage -le 15 ]; then
+  # Note: it's not important to give mkgraph.sh the lang directory with the
+  # matched topology (since it gets the topology file from the model).
+  utils/mkgraph.sh \
+    --self-loop-scale 1.0 data/lang_test_tgsmall \
+    $tree_dir $tree_dir/graph_tgsmall || exit 1;
+fi
+
+if [ $stage -le 16 ]; then
+  frames_per_chunk=$(echo $chunk_width | cut -d, -f1)
+  rm $dir/.error 2>/dev/null || true
+
+  for data in $test_sets; do
+    (
+      nspk=$(wc -l <data/${data}_hires/spk2utt)
+      steps/nnet3/decode.sh \
+          --acwt 1.0 --post-decode-acwt 10.0 \
+          --frames-per-chunk $frames_per_chunk \
+          --nj $nspk --cmd "$decode_cmd"  --num-threads 4 \
+          --online-ivector-dir exp/nnet3${nnet3_affix}/ivectors_${data}_hires \
+          $tree_dir/graph_tgsmall data/${data}_hires ${dir}/decode_tgsmall_${data} || exit 1
+      steps/lmrescore_const_arpa.sh --cmd "$decode_cmd" \
+        data/lang_test_{tgsmall,tglarge} \
+       data/${data}_hires ${dir}/decode_{tgsmall,tglarge}_${data} || exit 1
+    ) || touch $dir/.error &
+  done
+  wait
+  [ -f $dir/.error ] && echo "$0: there was a problem while decoding" && exit 1
+fi
+
+# Not testing the 'looped' decoding separately, because for
+# TDNN systems it would give exactly the same results as the
+# normal decoding.
+
+if $test_online_decoding && [ $stage -le 17 ]; then
+  # note: if the features change (e.g. you add pitch features), you will have to
+  # change the options of the following command line.
+  steps/online/nnet3/prepare_online_decoding.sh \
+    --mfcc-config conf/mfcc_hires.conf \
+    $lang exp/nnet3${nnet3_affix}/extractor ${dir} ${dir}_online
+
+  rm $dir/.error 2>/dev/null || true
+
+  for data in $test_sets; do
+    (
+      nspk=$(wc -l <data/${data}_hires/spk2utt)
+      # note: we just give it "data/${data}" as it only uses the wav.scp, the
+      # feature type does not matter.
+      steps/online/nnet3/decode.sh \
+        --acwt 1.0 --post-decode-acwt 10.0 \
+        --nj $nspk --cmd "$decode_cmd" \
+        $tree_dir/graph_tgsmall data/${data} ${dir}_online/decode_tgsmall_${data} || exit 1
+      steps/lmrescore_const_arpa.sh --cmd "$decode_cmd" \
+        data/lang_test_{tgsmall,tglarge} \
+       data/${data}_hires ${dir}_online/decode_{tgsmall,tglarge}_${data} || exit 1
+    ) || touch $dir/.error &
+  done
+  wait
+  [ -f $dir/.error ] && echo "$0: there was a problem while decoding" && exit 1
+fi
+
+
+exit 0;

egs/swbd/s5c/local/chain/run_tdnn.sh

@@ -1 +1 @@
-tuning/run_tdnn_7q.sh
+tuning/run_tdnn_7r.sh