speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

update

Browse Source
This commit is contained in:
嘉渊 2023-07-21 14:26:15 +08:00
parent 13a1ba2b1d
commit d273f7e126
2 changed files with 14 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
egs/callhome/eend_ola/local/gen_feats_scp.py
View File

@ -9,19 +9,17 @@ if __name__ == '__main__':
args = parser.parse_args()
root_path = args.root_path
out_path = args.out_path
datasets = ["train", "dev"]
split_num = args.split_num
for dataset in datasets:
with open(os.path.join(out_path, dataset, "feats.scp"), "w") as out_f:
for i in range(split_num):
idx = str(i + 1)
feature_file = os.path.join(root_path, dataset, "feature.scp.{}".format(idx))
label_file = os.path.join(root_path, dataset, "label.scp.{}".format(idx))
with open(feature_file) as ff, open(label_file) as fl:
ff_lines = ff.readlines()
fl_lines = fl.readlines()
for ff_line, fl_line in zip(ff_lines, fl_lines):
sample_name, f_path = ff_line.strip().split()
_, l_path = fl_line.strip().split()
out_f.write("{} {} {}\n".format(sample_name, f_path, l_path))
with open(os.path.join(out_path, "feats.scp"), "w") as out_f:
for i in range(split_num):
idx = str(i + 1)
feature_file = os.path.join(root_path, "feature.scp.{}".format(idx))
label_file = os.path.join(root_path, "label.scp.{}".format(idx))
with open(feature_file) as ff, open(label_file) as fl:
ff_lines = ff.readlines()
fl_lines = fl.readlines()
for ff_line, fl_line in zip(ff_lines, fl_lines):
sample_name, f_path = ff_line.strip().split()
_, l_path = fl_line.strip().split()
out_f.write("{} {} {}\n".format(sample_name, f_path, l_path))

4
egs/callhome/eend_ola/run_test.sh
View File

@ -101,8 +101,8 @@ if [ ${stage} -le 0 ] && [ ${stop_stage} -ge 0 ]; then
# --index JOB
mkdir -p ${data_dir}/ark_data/dump/simu_data/data/$dataset
python local/gen_feats_scp.py \
--root_path ${data_dir}/ark_data/dump/simu_data \
--out_path ${data_dir}/ark_data/dump/simu_data/data \
--root_path ${data_dir}/ark_data/dump/simu_data/$dataset \
--out_path ${data_dir}/ark_data/dump/simu_data/data/$dataset \
--split_num $nj
done
fi