mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-09-04 14:44:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update prepare_for_bpe_model.py

Browse Source
This commit is contained in:
jinzr 2023-11-23 11:20:37 +08:00
parent 7e798c21c7
commit d15605f660
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
egs/multi_zh_en/ASR/local/prepare_for_bpe_model.py
View File

@ -40,7 +40,7 @@ def get_args():
parser.add_argument( parser.add_argument(
"--text", "--text",
type=str, type=str,
help="WenetSpeech training transcript.", help="Training transcript.",
) )
return parser.parse_args() return parser.parse_args()
@ -56,7 +56,7 @@ def main():
transcript_path = lang_dir / "transcript_chars.txt" transcript_path = lang_dir / "transcript_chars.txt"
with open(text, "r", encoding="utf-8") as fin: with open(text, "r", encoding="utf-8") as fin:
with open(transcript_path, "w", encoding="utf-8") as fout: with open(transcript_path, "w+", encoding="utf-8") as fout:
for line in tqdm(fin): for line in tqdm(fin):
fout.write(tokenize_by_CJK_char(line) + "\n") fout.write(tokenize_by_CJK_char(line) + "\n")