安装配置kaldi的博客有很多,就不赘述了
进入$KALDI_ROOT(你的kaldi路径),到egs找到hkust,并把数据下好解压
修改run.sh中这一部分到你的数据路径:
# Data Preparation,
local/hkust_data_prep.sh /export/corpora/LDC/LDC2005S15/ /export/corpora/LDC/LDC2005T32/
然后运行,在hkust_data_prep.sh阶段报错:
Python module mmseg is not found. To install it, run tools/extra/install_mmseg.sh
根据提示到tools目录下运行 tools/extra/install_mmseg.sh
安装完成后输出:
Installation of mmseg finished successfully
Please source tools/env.sh in your path.sh to enable it
这里提醒我们记得在project里面的path.sh中source tools/env.sh,我们去看看示例里有没有这么做,到工程目录下:
cat path.sh
输出:
export KALDI_ROOT=`pwd`/../../..
[ -f $KALDI_ROOT/tools/env.sh ] && . $KALDI_ROOT/tools/env.sh
export PATH=$PWD/utils/:$KALDI_ROOT/tools/openfst/bin:$PWD:$PATH
[ ! -f $KALDI_ROOT/tools/config/common_path.sh ] && echo >&2 "The standard file $KALDI_ROOT/tools/config/common_path.sh is not present -> Exit!" && exit 1
. $KALDI_ROOT/tools/config/common_path.sh
export LC_ALL=C
可以看到第二行用点命令(等同于source)完成了这一操作,所以只需将. ./path.sh加入到run.sh中即可
hkust_prepare_dict.sh脚本运行时,报错g2p.py is not found. 解决方案:
cd $KALDI_ROOT/tools #
./extras/install_sequitur.sh
hkust_train_lms.sh运行train_lm.sh时,报错:train_lm.sh is not found. 解决方案:
cd $KALDI_ROOT/tools #
./extras/install_kaldi_lm.sh
在steps/make_mfcc_pitch_online.sh阶段报错:
queue.pl: Error submitting jobs to queue (return status was 32512)
queue log file is exp/make_mfcc/train/q/make_mfcc_pitch_train.log, command was qsub -v PATH -cwd -S /bin/bash -j y -l arch=64 -o exp/make_mfcc/train/q/make_mfcc_pitch_train.log -l mem_free=2G,ram_free=2G -t 1:10 /home/1/kaldi/egs/hkust/s5/exp/make_mfcc/train/q/make_mfcc_pitch_train.sh >>exp/make_mfcc/train/q/make_mfcc_pitch_train.log 2>&1
Output of qsub was: sh: 1: qsub: not found
解决方法,修改cmd.sh,将queue.pl --XXX改成run.pl即可