Kaldi语音识别技术

最新推荐文章于 2023-08-08 12:50:13 发布

花花少年

最新推荐文章于 2023-08-08 12:50:13 发布

阅读量2.4k

点赞数

分类专栏：深度学习文章标签：语音识别人工智能 Kaldi

本文链接：https://blog.csdn.net/m0_37605642/article/details/127469118

版权

深度学习专栏收录该内容

117 篇文章 77 订阅

订阅专栏

一、参考资料

如何用Kaldi做语音识别？

Kaldi官网

kaldi仓库

二、相关介绍

Kaldi是当前最流行的开源语音识别工具（Toolkit），旨在提供灵活且可扩展的组件，包括多种语音信号处理，语音识别，声纹识别和深度神经网络。Kaldi使用WFST来实现解码算法，主要由C++编写，在此之上使用bash和Python脚本做了一些工具。而实时识别系统的好坏，取决于语音识别的性能，语音识别包含“特征提取、声学模型、语言模型、解码器”等部分。Kaldi工具箱集成了几乎所有搭建语音识别器需要用到的工具。Kaldi支持GPU进行训练。

语音模型训练过程一般都非常复杂，对于同一数据集和模型，不同的人训练出来的结果可能会有很大差距。

2.1 TDNN

TDNN与xvector

[DL] 时延神经网络（TDNN）

**时延神经网络（TDNN）**来自1989年的论文《Phoneme recognition using time-delay neural networks》。原文中主要使用TDNN来识别音素，在识别"B", “D”, "G"三个浊音中得到98.5%的准确率，高于HMM的93.7%。

2.2 x-vector

x-vector的论文发表在ICASSP 2018，kaldi的核心开发者Daniel Povey也是这篇论文的作者之一，论文来链接如下：

X-VECTORS: ROBUST DNN EMBEDDINGS FOR SPEAKER RECOGNITION

X-VECTORS是当前声纹识别领域主流的baseline模型框架，得益于其网络中的statistics pooling层，X-VECTORS可接受任意长度的输入，转化为固定长度的特征表达；此外，在训练中引入了包含噪声和混响在内的数据增强策略，使得模型对于噪声和混响等干扰更加鲁棒。

三、关键步骤

保姆级kaldi语音识别（2）Linux系统Ubuntu20.04下开源语音识别工具kaldi配置

kaldi的编译安装与报错解决方法

3.1 准备工作

3.2 安装依赖

sudo apt-get update
sudo apt-get upgrade
sudo apt-get install git
sudo apt-get install bc
sudo apt-get install g++
sudo apt-get install zlib1g-dev make automake autoconf bzip2 libtool subversion
sudo apt-get install libatlas3-base

3.3 下载源码

git clone https://github.com/kaldi-asr/kaldi.git
cd kaldi/tools

3.4 检查依赖

用kaldi自带的脚本check_dependencies.sh来检测是否安装完所有必须的依赖工具，缺什么就安装什么。

extras/check_dependencies.sh

安装依赖

sudo apt-get install sox gfortran python2.7

安装MKL

extras/install_mkl.sh

extras/install_mkl.sh: Configuring ld runtime bindings
+ echo '/opt/intel/lib/intel64
/opt/intel/mkl/lib/intel64'
+ ldconfig
extras/install_mkl.sh: MKL package intel-mkl-64bit-2019.2-057 was successfully installed

安装openfst

重新配置configure

cd openfst-1.7.2/
sudo ./configure

编译openfst

cd ..
sudo make openfst

3.5 （再次）检查依赖

extras/check_dependencies.sh

extras/check_dependencies.sh: all OK.

3.6 编译tools

依赖安装成功，说明编译所需的工具和环境都配置好了，接下来就可以编译tools。

cd kaldi/tools
sudo make -j 8

3.7 编译src

cd src
./configure --shared

make depend -j

make -j 12

mkdir /mnt/d/MyDocuments/cache/kaldi/src/lib
The version of configure script matches kaldi.mk version. Good.
make -C base
make[1]: Entering directory '/mnt/d/MyDocuments/cache/kaldi/src/base'
...
...
...
make[1]: Leaving directory '/mnt/d/MyDocuments/cache/kaldi/src/latbin'
Done

3.8 运行yesno样例

cd ..
cd egs/yesno/s5
./run.sh

Preparing train and test data
Dictionary preparation succeeded
utils/prepare_lang.sh --position-dependent-phones false data/local/dict <SIL> data/local/lang data/lang
Checking data/local/dict/silence_phones.txt ...
--> reading data/local/dict/silence_phones.txt
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> data/local/dict/silence_phones.txt is OK

Checking data/local/dict/optional_silence.txt ...
--> reading data/local/dict/optional_silence.txt
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> data/local/dict/optional_silence.txt is OK

Checking data/local/dict/nonsilence_phones.txt ...
--> reading data/local/dict/nonsilence_phones.txt
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> data/local/dict/nonsilence_phones.txt is OK

Checking disjoint: silence_phones.txt, nonsilence_phones.txt
--> disjoint property is OK.

Checking data/local/dict/lexicon.txt
--> reading data/local/dict/lexicon.txt
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> data/local/dict/lexicon.txt is OK

Checking data/local/dict/extra_questions.txt ...
--> data/local/dict/extra_questions.txt is empty (this is OK)
--> SUCCESS [validating dictionary directory data/local/dict]

**Creating data/local/dict/lexiconp.txt from data/local/dict/lexicon.txt
fstaddselfloops data/lang/phones/wdisambig_phones.int data/lang/phones/wdisambig_words.int
prepare_lang.sh: validating output directory
utils/validate_lang.pl data/lang
Checking existence of separator file
separator file data/lang/subword_separator.txt is empty or does not exist, deal in word case.
Checking data/lang/phones.txt ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> data/lang/phones.txt is OK

Checking words.txt: #0 ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> data/lang/words.txt is OK

Checking disjoint: silence.txt, nonsilence.txt, disambig.txt ...
--> silence.txt and nonsilence.txt are disjoint
--> silence.txt and disambig.txt are disjoint
--> disambig.txt and nonsilence.txt are disjoint
--> disjoint property is OK

Checking sumation: silence.txt, nonsilence.txt, disambig.txt ...
--> found no unexplainable phones in phones.txt

Checking data/lang/phones/context_indep.{txt, int, csl} ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> 1 entry/entries in data/lang/phones/context_indep.txt
--> data/lang/phones/context_indep.int corresponds to data/lang/phones/context_indep.txt
--> data/lang/phones/context_indep.csl corresponds to data/lang/phones/context_indep.txt
--> data/lang/phones/context_indep.{txt, int, csl} are OK

Checking data/lang/phones/nonsilence.{txt, int, csl} ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> 2 entry/entries in data/lang/phones/nonsilence.txt
--> data/lang/phones/nonsilence.int corresponds to data/lang/phones/nonsilence.txt
--> data/lang/phones/nonsilence.csl corresponds to data/lang/phones/nonsilence.txt
--> data/lang/phones/nonsilence.{txt, int, csl} are OK

Checking data/lang/phones/silence.{txt, int, csl} ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> 1 entry/entries in data/lang/phones/silence.txt
--> data/lang/phones/silence.int corresponds to data/lang/phones/silence.txt
--> data/lang/phones/silence.csl corresponds to data/lang/phones/silence.txt
--> data/lang/phones/silence.{txt, int, csl} are OK

Checking data/lang/phones/optional_silence.{txt, int, csl} ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> 1 entry/entries in data/lang/phones/optional_silence.txt
--> data/lang/phones/optional_silence.int corresponds to data/lang/phones/optional_silence.txt
--> data/lang/phones/optional_silence.csl corresponds to data/lang/phones/optional_silence.txt
--> data/lang/phones/optional_silence.{txt, int, csl} are OK

Checking data/lang/phones/disambig.{txt, int, csl} ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> 2 entry/entries in data/lang/phones/disambig.txt
--> data/lang/phones/disambig.int corresponds to data/lang/phones/disambig.txt
--> data/lang/phones/disambig.csl corresponds to data/lang/phones/disambig.txt
--> data/lang/phones/disambig.{txt, int, csl} are OK

Checking data/lang/phones/roots.{txt, int} ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> 3 entry/entries in data/lang/phones/roots.txt
--> data/lang/phones/roots.int corresponds to data/lang/phones/roots.txt
--> data/lang/phones/roots.{txt, int} are OK

Checking data/lang/phones/sets.{txt, int} ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> 3 entry/entries in data/lang/phones/sets.txt
--> data/lang/phones/sets.int corresponds to data/lang/phones/sets.txt
--> data/lang/phones/sets.{txt, int} are OK

Checking data/lang/phones/extra_questions.{txt, int} ...
Checking optional_silence.txt ...
--> reading data/lang/phones/optional_silence.txt
--> data/lang/phones/optional_silence.txt is OK

Checking disambiguation symbols: #0 and #1
--> data/lang/phones/disambig.txt has "#0" and "#1"
--> data/lang/phones/disambig.txt is OK

Checking topo ...

Checking word-level disambiguation symbols...
--> data/lang/phones/wdisambig.txt exists (newer prepare_lang.sh)
Checking data/lang/oov.{txt, int} ...
--> text seems to be UTF-8 or ASCII, checking whitespaces
--> text contains only allowed whitespaces
--> 1 entry/entries in data/lang/oov.txt
--> data/lang/oov.int corresponds to data/lang/oov.txt
--> data/lang/oov.{txt, int} are OK

--> data/lang/L.fst is olabel sorted
--> data/lang/L_disambig.fst is olabel sorted
--> SUCCESS [validating lang directory data/lang]
Preparing language models for test
arpa2fst --disambig-symbol=#0 --read-symbol-table=data/lang_test_tg/words.txt input/task.arpabo data/lang_test_tg/G.fst
LOG (arpa2fst[5.5.1056~1-f6f4c]:Read():arpa-file-parser.cc:94) Reading \data\ section.
LOG (arpa2fst[5.5.1056~1-f6f4c]:Read():arpa-file-parser.cc:149) Reading \1-grams: section.
LOG (arpa2fst[5.5.1056~1-f6f4c]:RemoveRedundantStates():arpa-lm-compiler.cc:359) Reduced num-states from 1 to 1
fstisstochastic data/lang_test_tg/G.fst
1.20397 1.20397
Succeeded in formatting data.
steps/make_mfcc.sh --nj 1 data/train_yesno exp/make_mfcc/train_yesno mfcc
utils/validate_data_dir.sh: WARNING: you have only one speaker.  This probably a bad idea.
   Search for the word 'bold' in http://kaldi-asr.org/doc/data_prep.html
   for more information.
utils/validate_data_dir.sh: Successfully validated data-directory data/train_yesno
steps/make_mfcc.sh: [info]: no segments file exists: assuming wav.scp indexed by utterance.
steps/make_mfcc.sh: Succeeded creating MFCC features for train_yesno
steps/compute_cmvn_stats.sh data/train_yesno exp/make_mfcc/train_yesno mfcc
Succeeded creating CMVN stats for train_yesno
fix_data_dir.sh: kept all 31 utterances.
fix_data_dir.sh: old files are kept in data/train_yesno/.backup
steps/make_mfcc.sh --nj 1 data/test_yesno exp/make_mfcc/test_yesno mfcc
utils/validate_data_dir.sh: WARNING: you have only one speaker.  This probably a bad idea.
   Search for the word 'bold' in http://kaldi-asr.org/doc/data_prep.html
   for more information.
utils/validate_data_dir.sh: Successfully validated data-directory data/test_yesno
steps/make_mfcc.sh: [info]: no segments file exists: assuming wav.scp indexed by utterance.
steps/make_mfcc.sh: It seems not all of the feature files were successfully procesed (29 != 31); consider using utils/fix_data_dir.sh data/test_yesno
steps/make_mfcc.sh: Less than 95% the features were successfully generated. Probably a serious error.
steps/compute_cmvn_stats.sh data/test_yesno exp/make_mfcc/test_yesno mfcc
Succeeded creating CMVN stats for test_yesno
fix_data_dir.sh: kept 29 utterances out of 31
fix_data_dir.sh: old files are kept in data/test_yesno/.backup
steps/train_mono.sh --nj 1 --cmd utils/run.pl --totgauss 400 data/train_yesno data/lang exp/mono0a
steps/train_mono.sh: Initializing monophone system.
steps/train_mono.sh: Compiling training graphs
steps/train_mono.sh: Aligning data equally (pass 0)
steps/train_mono.sh: Pass 1
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 2
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 3
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 4
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 5
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 6
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 7
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 8
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 9
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 10
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 11
steps/train_mono.sh: Pass 12
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 13
steps/train_mono.sh: Pass 14
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 15
steps/train_mono.sh: Pass 16
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 17
steps/train_mono.sh: Pass 18
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 19
steps/train_mono.sh: Pass 20
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 21
steps/train_mono.sh: Pass 22
steps/train_mono.sh: Pass 23
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 24
steps/train_mono.sh: Pass 25
steps/train_mono.sh: Pass 26
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 27
steps/train_mono.sh: Pass 28
steps/train_mono.sh: Pass 29
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 30
steps/train_mono.sh: Pass 31
steps/train_mono.sh: Pass 32
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 33
steps/train_mono.sh: Pass 34
steps/train_mono.sh: Pass 35
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 36
steps/train_mono.sh: Pass 37
steps/train_mono.sh: Pass 38
steps/train_mono.sh: Aligning data
steps/train_mono.sh: Pass 39
steps/diagnostic/analyze_alignments.sh --cmd utils/run.pl data/lang exp/mono0a
run.pl: job failed, log is in exp/mono0a/log/analyze_alignments.log
steps/diagnostic/analyze_alignments.sh: analyze_phone_length_stats.py failed, but ignoring the error (it's just for diagnostics)
steps/diagnostic/analyze_alignments.sh: see stats in exp/mono0a/log/analyze_alignments.log
1 warnings in exp/mono0a/log/update.*.log
exp/mono0a: nj=1 align prob=-81.88 over 0.05h [retry=0.0%, fail=0.0%] states=11 gauss=371
steps/train_mono.sh: Done training monophone system in exp/mono0a
tree-info exp/mono0a/tree
tree-info exp/mono0a/tree
fstminimizeencoded
fstpushspecial
fsttablecompose data/lang_test_tg/L_disambig.fst data/lang_test_tg/G.fst
fstdeterminizestar --use-log=true
fstisstochastic data/lang_test_tg/tmp/LG.fst
0.534295 0.533859
[info]: LG not stochastic.
fstcomposecontext --context-size=1 --central-position=0 --read-disambig-syms=data/lang_test_tg/phones/disambig.int --write-disambig-syms=data/lang_test_tg/tmp/disambig_ilabels_1_0.int data/lang_test_tg/tmp/ilabels_1_0.11190 data/lang_test_tg/tmp/LG.fst
fstisstochastic data/lang_test_tg/tmp/CLG_1_0.fst
0.534295 0.533859
[info]: CLG not stochastic.
make-h-transducer --disambig-syms-out=exp/mono0a/graph_tgpr/disambig_tid.int --transition-scale=1.0 data/lang_test_tg/tmp/ilabels_1_0 exp/mono0a/tree exp/mono0a/final.mdl
fstrmsymbols exp/mono0a/graph_tgpr/disambig_tid.int
fsttablecompose exp/mono0a/graph_tgpr/Ha.fst data/lang_test_tg/tmp/CLG_1_0.fst
fstdeterminizestar --use-log=true
fstrmepslocal
fstminimizeencoded
fstisstochastic exp/mono0a/graph_tgpr/HCLGa.fst
0.5342 -0.000422432
HCLGa is not stochastic
add-self-loops --self-loop-scale=0.1 --reorder=true exp/mono0a/final.mdl exp/mono0a/graph_tgpr/HCLGa.fst
steps/decode.sh --nj 1 --cmd utils/run.pl exp/mono0a/graph_tgpr data/test_yesno exp/mono0a/decode_test_yesno
decode.sh: feature type is delta
steps/diagnostic/analyze_lats.sh --cmd utils/run.pl exp/mono0a/graph_tgpr exp/mono0a/decode_test_yesno
run.pl: job failed, log is in exp/mono0a/decode_test_yesno/log/analyze_alignments.log
local/score.sh --cmd utils/run.pl data/test_yesno exp/mono0a/graph_tgpr exp/mono0a/decode_test_yesno
local/score.sh: scoring with word insertion penalty=0.0,0.5,1.0
%WER 0.00 [ 0 / 232, 0 ins, 0 del, 0 sub ] exp/mono0a/decode_test_yesno/wer_10_0.0

3.9 运行成功

%WER 0.00 [ 0 / 232, 0 ins, 0 del, 0 sub ] exp/mono0a/decode_test_yesno/wer_10_0.0

四、说话人识别系统

使用kaldi中的x-vector在aishell数据库上建立说话人识别系统

使用kaldi中的x-vector在aishell数据库上建立说话人识别系统。

整个系统分为三部分：

前端预处理部分，主要包括mfcc特征提取，VAD，数据扩充（增加混响、增加不同类型的噪声）等；
基于TDNN的特征提取器，该结构生成说话人表征，说话人表征也称为“embedding、x-vector”
后处理部分，对于说话人表征，采用LDA进行降维并训练PLDA模型对测试进行打分。

五、FAQ

Q：`c++: fatal error: Killed signal terminated program cc1plus`

C++: fatal error: Killed signal terminated program cc1plus的问题解决

c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: arpa-to-const-arpa.o] Error 1
make[1]: Leaving directory '/mnt/d/MyDocuments/cache/kaldi/src/lmbin'
make: *** [Makefile:168: lmbin] Error 2
make: *** Waiting for unfinished jobs....
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: apply-cmvn.o] Error 1
make[1]: *** Waiting for unfinished jobs....
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: feat-to-dim.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: ivector-subtract-global-mean.o] Error 1
make[1]: *** Waiting for unfinished jobs....
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: compose-transforms.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: minimize-lattice.o] Error 1
make[1]: *** Waiting for unfinished jobs....
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: ivector-mean.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: compute-and-process-kaldi-pitch-feats.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: paste-vectors.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: extend-transform-dim.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: extract-feature-segments.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: multiply-vectors.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: paste-feats.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: compute-cmvn-stats-two-channel.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: subset-feats.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: compute-fbank-feats.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: fmpe-acc-stats.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: compute-vad-from-frame-likes.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: fmpe-apply-transform.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: compute-mfcc-feats.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: agglomerative-cluster.o] Error 1
c++: fatal error: Killed signal terminated program cc1plus
compilation terminated.
make[1]: *** [<builtin>: determinize-lattice-pruned.o] Error 1
make[1]: Leaving directory '/mnt/d/MyDocuments/cache/kaldi/src/featbin'
make: *** [Makefile:168: featbin] Error 2
make[1]: Leaving directory '/mnt/d/MyDocuments/cache/kaldi/src/ivectorbin'
make: *** [Makefile:168: ivectorbin] Error 2
ar -cr kaldi-nnet.a nnet-nnet.o nnet-component.o nnet-loss.o nnet-pdf-prior.o nnet-randomizer.o
ranlib kaldi-nnet.a
c++ -shared -o libkaldi-nnet.so -Wl,--as-needed  -Wl,-soname=libkaldi-nnet.so,--whole-archive kaldi-nnet.a -Wl,--no-whole-archive -Wl,-rpath=/mnt/d/MyDocuments/cache/kaldi/tools/openfst-1.7.2/lib -rdynamic   -Wl,-rpath=/mnt/d/MyDocuments/cache/kaldi/src/lib  ../cudamatrix/libkaldi-cudamatrix.so  ../hmm/libkaldi-hmm.so  ../tree/libkaldi-tree.so  ../util/libkaldi-util.so  ../matrix/libkaldi-matrix.so  ../base/libkaldi-base.so /mnt/d/MyDocuments/cache/kaldi/tools/openfst-1.7.2/lib/libfst.so -L/opt/intel/mkl/lib/intel64 -Wl,-rpath=/opt/intel/mkl/lib/intel64 -l:libmkl_intel_lp64.so -l:libmkl_core.so -l:libmkl_sequential.so -ldl -lpthread -lm -lm -lpthread -ldl
ln -sf /mnt/d/MyDocuments/cache/kaldi/src/nnet/libkaldi-nnet.so /mnt/d/MyDocuments/cache/kaldi/src/lib/libkaldi-nnet.so
make[1]: Leaving directory '/mnt/d/MyDocuments/cache/kaldi/src/nnet'
make[1]: Leaving directory '/mnt/d/MyDocuments/cache/kaldi/src/lat'
make: *** [Makefile:164: lat] Error 2

错误原因：
内存不足，导致编译中断

解决办法：
增加进程数量，增加make编译时的进程数量
make -j 12

其他办法：
（1）增加内存
（2）增加SWAP交换空间
（3）减少进程数量，减少make编译时的进程数量

花花少年

关注

0
点赞
踩
11

收藏

觉得还不错? 一键收藏
打赏
1
评论
Kaldi语音识别技术

Kaldi是当前最流行的开源语音识别工具（Toolkit），旨在提供灵活且可扩展的组件，包括多种语音信号处理，语音识别，声纹识别和深度神经网络。Kaldi使用WFST来实现解码算法，主要由C++编写，在此之上使用bash和Python脚本做了一些工具。而实时识别系统的好坏，取决于语音识别的性能，语音识别包含“特征提取、声学模型、语言模型、解码器”等部分。Kaldi工具箱集成了几乎所有搭建语音识别器需要用到的工具。Kaldi支持GPU进行训练。
复制链接

扫一扫