From be00b3df2c04a81e216eb582a0ab89afb37d855c Mon Sep 17 00:00:00 2001
From: sathvik udupa <sathvikudupa66@gmail.com>
Date: Wed, 19 Apr 2023 18:56:26 +0530
Subject: [PATCH] modify cut json with trim supervision

---
 egs/mucs/ASR/local/compute_fbank_mucs.py | 5 +++++
 egs/mucs/ASR/prepare.sh                  | 8 ++++----
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/egs/mucs/ASR/local/compute_fbank_mucs.py b/egs/mucs/ASR/local/compute_fbank_mucs.py
index ef2ddeb46..fa486cd74 100755
--- a/egs/mucs/ASR/local/compute_fbank_mucs.py
+++ b/egs/mucs/ASR/local/compute_fbank_mucs.py
@@ -104,6 +104,8 @@ def compute_fbank_mucs(
 
     with get_executor() as ex:  # Initialize the executor only once.
         for partition, m in manifests.items():
+            # print(m["recordings"])
+            # exit()
             cuts_filename = f"{prefix}_cuts_{partition}.{suffix}"
             if (output_dir / cuts_filename).is_file():
                 logging.info(f"{partition} already exists - skipping.")
@@ -128,6 +130,9 @@ def compute_fbank_mucs(
                 executor=ex,
                 storage_type=LilcomChunkyWriter,
             )
+            cut_set = cut_set.trim_to_supervisions(
+            keep_overlapping=False, min_duration=None
+        )
             cut_set.to_file(output_dir / cuts_filename)
 
 
diff --git a/egs/mucs/ASR/prepare.sh b/egs/mucs/ASR/prepare.sh
index b8a09e39e..37ed117c6 100755
--- a/egs/mucs/ASR/prepare.sh
+++ b/egs/mucs/ASR/prepare.sh
@@ -6,8 +6,8 @@ export PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python
 set -eou pipefail
 
 nj=60
-stage=8
-stop_stage=8
+stage=3
+stop_stage=3
 
 # We assume dl_dir (download dir) contains the following
 # directories and files. If not, they will be downloaded
@@ -97,12 +97,12 @@ if [ $stage -le 3 ] && [ $stop_stage -ge 3 ]; then
     touch data/fbank/.mucs.done
   fi
 
-  
+  # exit
 
   if [ ! -e data/fbank/.mucs-validated.done ]; then
     log "Validating data/fbank for mucs"
     parts=(
-      train
+      train,
       test
     )
     for part in ${parts[@]}; do