some bugfixes

egs/fisher_swbd/ASR/local/normalize_and_filter_supervisions.py

@@ -4,10 +4,9 @@ import argparse
 import re
 from typing import Tuple
 
-from tqdm import tqdm
+from lhotse import SupervisionSegment, SupervisionSet
-
-from lhotse import SupervisionSet, SupervisionSegment
 from lhotse.serialization import load_manifest_lazy_or_eager
+from tqdm import tqdm
 
 
 def get_args():
@@ -19,19 +18,14 @@ def get_args():
 
 # fmt: off
 class FisherSwbdNormalizer:
+    """Note: the functions "normalize" and "keep" implement the logic
+    similar to Kaldi's data prep scripts for Fisher and SWBD: One
+    notable difference is that we don't change [cough], [lipsmack],
+    etc. to [noise].  We also don't implement all the edge cases of
+    normalization from Kaldi (hopefully won't make too much
+    difference).
+
     """
-    Note: the functions "normalize" and "keep" implement the logic similar to
-    Kaldi's data prep scripts for Fisher:
-      https://github.com/kaldi-asr/kaldi/blob/master/egs/fisher_swbd/s5/local/fisher_data_prep.sh
-    and for SWBD:
-      https://github.com/kaldi-asr/kaldi/blob/master/egs/fisher_swbd/s5/local/swbd1_data_prep.sh
-
-    One notable difference is that we don't change [cough], [lipsmack], etc. to [noise]. 
-    We also don't implement all the edge cases of normalization from Kaldi 
-    (hopefully won't make too much difference).
-    """
-
-
     def __init__(self) -> None:
 
         self.remove_regexp_before = re.compile(

egs/fisher_swbd/ASR/prepare.sh

@@ -28,8 +28,8 @@ stop_stage=500
 #     - noise
 #     - speech
 
-dl_dir=$PWD/download
+dl_dir=/mnt/dsk2
-mkdir -p $dl_dir
+#mkdir -p $dl_dir
 
 . shared/parse_options.sh || exit 1
 
@@ -62,25 +62,25 @@ if [ $stage -le 0 ] && [ $stop_stage -ge 0 ]; then
   #
 
   # TODO: remove
-  LDC_ROOT=/nas/data4/DATA
+#  LDC_ROOT=/nas/data4/DATA
-  for pkg in LDC2004S13 LDC2004T19 LDC2005S13 LDC2005T19 LDC97S62 LDC2002S09 LDC2002T43; do
+#  for pkg in LDC2004S13 LDC2004T19 LDC2005S13 LDC2005T19 LDC97S62 LDC2002S09 LDC2002T43; do
-    ln -sfv $LDC_ROOT/$pkg $dl_dir/
+#    ln -sfv $LDC_ROOT/$pkg $dl_dir/
-  done
+#  done
 
   # If you have pre-downloaded it to /path/to/musan,
   # you can create a symlink
   #
   #   ln -sfv /path/to/musan $dl_dir/
   #
-  if [ ! -d $dl_dir/musan ]; then
+#  if [ ! -d $dl_dir/musan ]; then
-    lhotse download musan $dl_dir
+#    lhotse download musan $dl_dir
-  fi
+#  fi
 fi
 
 if [ $stage -le 1 ] && [ $stop_stage -ge 1 ] ; then
   log "Stage 1: Prepare Fisher manifests"
-  mkdir -p data/manifests/fisher
+#  mkdir -p data/manifests/fisher
-  lhotse prepare fisher-english --absolute-paths 1 $dl_dir data/manifests/fisher
+#  lhotse prepare fisher-english --absolute-paths 1 $dl_dir data/manifests/fisher
   local/normalize_and_filter_supervisions.py data/manifests/fisher/supervisions.jsonl.gz data/manifests/supervisions_fisher.jsonl.gz
   cp data/manifests/fisher/recordings.jsonl.gz data/manifests/recordings_fisher.jsonl.gz
   gzip -d data/manifests/supervisions_fisher.jsonl.gz
@@ -91,10 +91,10 @@ if [ $stage -le 2 ] && [ $stop_stage -ge 2 ]; then
   log "Stage 2: Prepare SWBD manifests"
   mkdir -p data/manifests/swbd
   lhotse prepare switchboard --absolute-paths 1 --omit-silence $dl_dir/LDC97S62 data/manifests/swbd
-  python3 local/normalize_and_filter_supervisions.py data/manifests/swbd/swbd_supervisions_all.jsonl.gz data/manifests/supervisions_swbd.jsonl.gz
+  python3 local/normalize_and_filter_supervisions.py data/manifests/swbd/swbd_supervisions.jsonl data/manifests/supervisions_swbd.jsonl
-  cp data/manifests/swbd/swbd_recordings_all.jsonl.gz data/manifests/recordings_swbd.jsonl.gz
+  cp data/manifests/swbd/swbd_recordings.jsonl data/manifests/recordings_swbd.jsonl
-  gzip -d data/manifests/supervisions_swbd.jsonl.gz
+#  gzip -d data/manifests/supervisions_swbd.jsonl.gz
-  gzip -d data/manifests/recordings_swbd.jsonl.gz
+#  gzip -d data/manifests/recordings_swbd.jsonl.gz
 fi
 
 if [ $stage -le 3 ] && [ $stop_stage -ge 3 ]; then