锐英源软件
第一信赖

精通

英语

开源

擅长

开发

培训

胸怀四海 

第一信赖

当前位置:锐英源 / 开源技术 / 语音识别开源 / kaldi清华模型训练中遇到core错误
服务方向
人工智能数据处理
人工智能培训
kaldi数据准备
小语种语音识别
语音识别标注
语音识别系统
语音识别转文字
kaldi开发技术服务
软件开发
运动控制卡上位机
机械加工软件
软件开发培训
Java 安卓移动开发
VC++
C#软件
汇编和破解
驱动开发
联系方式
固话:0371-63888850
手机:138-0381-0136
Q Q:396806883
微信:ryysoft

kaldi清华模型训练中遇到core错误


-------tri2b------- est-lda

steps/train_lda_mllt.sh --cmd run.pl --splice-opts --left-context=3 --right-context=3 2500 15000 data/mfcc/train data/lang exp/tri1_ali exp/tri2b
steps/train_lda_mllt.sh: Accumulating LDA statistics.
steps/train_lda_mllt.sh: line 116:  8217 Aborted                 (core dumped) est-lda --write-full-matrix=$dir/full.mat --dim=$dim $dir/0.mat $dir/lda.*.acc 2> $dir/log/lda_est.log

est-lda --write-full-matrix=exp/tri2b/full.mat --dim=40 exp/tri2b/0.mat exp/tri2b/lda.1.acc
ASSERTION_FAILED (est-lda[5.4.164~2-13ed]:Estimate():lda-estimate.cc:91) : 'target_dim <= Dim() && (target_dim < NumClasses() || opts.allow_large_dim)'

[ Stack-Trace: ]

kaldi::MessageLogger::HandleMessage(kaldi::LogMessageEnvelope const&, char const*)
kaldi::MessageLogger::~MessageLogger()
kaldi::KaldiAssertFailure_(char const*, char const*, int, char const*)
kaldi::LdaEstimate::Estimate(kaldi::LdaEstimateOptions const&, kaldi::Matrix<float>*, kaldi::Matrix<float>*) const
main
__libc_start_main
_start

exp/tri2b/0.mat这个文件没有,另外一个有.

run界面上有输出:
mkgraph.sh: expected data/graph/lang/L.fst to exist,这个提示在有文件时也报.
tri阶段

WARNING (gmm-init-model[5.4.164~2-13ed]:InitAmGmm():gmm-init-model.cc:55) Tree has pdf-id 10 with no stats; corresponding phone list: 11
** The warnings above about 'no stats' generally mean you have phones **
** (or groups of phones) in your phone set that had no corresponding data. **
** You should probably figure out whether something went wrong, **
** or whether your data just doesn't happen to have examples of those **
** phones. **

但是原来模型识别输出里有。

SPLICE
Splice在网上说它的也很少,这里采用Kaldi的原话:
Frame splicing (e.g. splicing nine consecutive frames together) is typically done to the raw MFCC features prior to LDA. The program splice-feats does this. A typical line from a script that uses this is the following:
feats="ark:splice-feats scp:data/train.scp ark:- | transform-feats $dir/0.mat ark:- ark:- |"

  • 1

and the “feats” variable would later be used as an rspecifier (c.f. Specifying Table formats: wspecifiers and rspecifiers) by some program that needs to read features. In this example we don’t specify the number of frames to splice together because we are using the defaults (–left-context=4, –right-context=4, or 9 frames in total).
原始mfcc到lda转换。

est-lda [options] <lda-matrix-out> <lda-acc-1> <lda-acc-2>
target_dim是40,
Dim()是first_acc_的维度exp/tri2b/lda.1.acc的
NumClasses first_acc_.NumRows()

还是让wav搞短点。

后来认真看代码改好了。

友情链接
版权所有 Copyright(c)2004-2021 锐英源软件
公司注册号:410105000449586 豫ICP备08007559号 最佳分辨率 1024*768
地址:郑州大学北校区院(文化路97号院)内