diff --git a/examples/hi_xiaowen/s0/run.sh b/examples/hi_xiaowen/s0/run.sh
index a26d16e..32b3257 100755
--- a/examples/hi_xiaowen/s0/run.sh
+++ b/examples/hi_xiaowen/s0/run.sh
@@ -7,13 +7,15 @@ stage=3
 stop_stage=3
 num_keywords=2
 
+config=conf/tcn.yaml            
 norm_mean=true
 norm_var=true
 gpus="0,1"
 
+dir=exp/ds_tcn
 
 num_average=30
-checkpoint=$dir/avg_${num_average}.pt
+checkpoint=
 score_checkpoint=$dir/avg_${num_average}.pt
 
 
@@ -94,8 +96,8 @@ if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
   mkdir -p $result_dir
   python kws/bin/score_longwav.py \
     --config $dir/config.yaml \
-    --test_data data/test/test_data.list \
-    --batch_size 5 \
+    --test_data data/test/data.list \
+    --batch_size 256 \
     --checkpoint $score_checkpoint \
     --score_file_dir $result_dir  \
     --num_keywords $num_keywords  \
@@ -104,7 +106,7 @@ if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then
   for keyword in 0 1; do
     python kws/bin/compute_det_longwav.py \
       --keyword $keyword \
-      --test_data data/test/test_data.list \
+      --test_data data/test/data.list \
       --score_file $result_dir/score_longwav.${keyword}.txt \
       --stats_file $result_dir/stats_longwav.${keyword}.txt
   done
diff --git a/kws/bin/compute_det_longwav.py b/kws/bin/compute_det_longwav.py
index 5d26afb..02c827e 100644
--- a/kws/bin/compute_det_longwav.py
+++ b/kws/bin/compute_det_longwav.py
@@ -68,9 +68,6 @@ if __name__ == '__main__':
     keyword_table, filler_table, filler_duration = load_label_and_score(
         args.keyword, args.test_data, args.score_file)
     print('Filler total duration Hours: {}'.format(filler_duration / 3600.0))
-    # print('keyword_table.size = ', len(keyword_table))
-    # print('filler_table.size = ', len(filler_table))
-    # print('filler_duration = ', filler_duration)
     with open(args.stats_file, 'w', encoding='utf8') as fout:
         threshold = 0.0
         while threshold <= 1.0:
@@ -88,9 +85,9 @@ if __name__ == '__main__':
                 while i < len(score_list):
                     if score_list[i] >= threshold:
                         num_false_alarm += 1
-                        i += 1
-                    else:
                         i += window_shift
+                    else:
+                        i += 1
             if len(keyword_table) != 0 :
                 false_reject_rate = num_false_reject / len(keyword_table)
             num_false_alarm = max(num_false_alarm, 1e-6)