Aishell tdnn training: fix_data_dir.sh: no utterances remained: not proceeding further.

[Peter]

I changed the nj(numbers of job) from 10 to 20, would it affect tdnn training ? 

There is no problem when i train mono, tri1, tri2, and up to tri5, but when i tried to train nnet3 tdnn, it just output the following two lines log:

local/nnet3/run_ivector_common.sh: preparing directory for low-resolution speed-perturbed data (for alignment)

fix_data_dir.sh: no utterances remained: not proceeding further.

Regards,

Peter

[软件开发工作经验]

dataset is maybe changed by nj,I once changed nj=1 to train,so if your dataset is small,don't let nj=20.

------------------ 原始邮件 ------------------

发件人: "kaldi-help" <petertse...@gmail.com>;

发送时间: 2021年10月9日(星期六) 中午12:00

收件人: "kaldi-help"<kaldi...@googlegroups.com>;

主题: [kaldi-help] Aishell tdnn training: fix_data_dir.sh: no utterances remained: not proceeding further.

--
Go to http://kaldi-asr.org/forums.html to find out how to join the kaldi-help group
---
You received this message because you are subscribed to the Google Groups "kaldi-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to kaldi-help+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/kaldi-help/4ac70570-c513-4424-b3d2-9fa335127ff2n%40googlegroups.com.