You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
4
Traceback (most recent call last):
File "local/extract_meta.py", line 102, in
main()
File "local/extract_meta.py", line 98, in main
meta_analysis(args.input_json, args.output_dir)
File "local/extract_meta.py", line 47, in meta_analysis
if json_data['audio'] is not None:
KeyError: 'audio'
The text was updated successfully, but these errors were encountered:
1 我先是按照下载方式中通过魔塔社区下载的数据集,它是自动就给分好train、test等音频数据集,还有一个json文件与切分数据集的名称对应
2 stage 0这一步骤(wenetspeech_data_prep.sh),准备数据部分出现问题,我是把下载数据的json文件放在对应的路径下(但我总感觉不是这个json文件,这个只是对应划分训练集的),执行到下面 3 这个代码时候出现 4 的报错。求解!!没找到有写训练流程的教程。
3 # Files to be created:
wav.scp text segments utt2dur
python3 local/extract_meta.py
$wenetspeech_dir/WenetSpeech.json $corpus_dir || exit 1;
4
Traceback (most recent call last):
File "local/extract_meta.py", line 102, in
main()
File "local/extract_meta.py", line 98, in main
meta_analysis(args.input_json, args.output_dir)
File "local/extract_meta.py", line 47, in meta_analysis
if json_data['audio'] is not None:
KeyError: 'audio'
The text was updated successfully, but these errors were encountered: