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的nnet2模型训练bad dimension和Assertion failed错误
联系方式
固话:0371-63888850
手机:138-0381-0136
Q Q:396806883
微信:ryysoft
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

锐英源精品原创,禁止全文或局部转载,禁止任何形式的非法使用,侵权必究


kaldi的nnet2模型训练bad dimension和Assertion failed错误


背景

锐英源软件在训练nnet2模型时遇到类似标题的错误,根据环境分析解决了问题,如果有类似问题的朋友可以联系锐英源软件,因为资料太少,处理这类问题很头疼,微信手机13803810136。

 

日志

WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_019 has bad dimension 40x91 versus feat dim 280
WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_020 has bad dimension 40x91 versus feat dim 280
WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_021 has bad dimension 40x91 versus feat dim 280
WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_022 has bad dimension 40x91 versus feat dim 280
WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_023 has bad dimension 40x91 versus feat dim 280
WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_024 has bad dimension 40x91 versus feat dim 280
WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_025 has bad dimension 40x91 versus feat dim 280
LOG (apply-cmvn-online[5.5.839~8-0c6a]:main():apply-cmvn-online.cc:133) Applied online CMVN to 7138 files, or 2708989 frames.
WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_026 has bad dimension 40x91 versus feat dim 280
WARNING (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:110) Transform matrix for utterance A74_027 has bad dimension 40x91 versus feat dim 280
LOG (transform-feats[5.5.839~8-0c6a]:main():transform-feats.cc:161) Applied transform to 0 utterances; 7138 had errors.
LOG (gmm-global-init-from-feats[5.5.839~8-0c6a]:main():gmm-global-init-from-feats.cc:144) Reading features (will keep 400000 frames.)
WARNING (gmm-global-init-from-feats[5.5.839~8-0c6a]:main():gmm-global-init-from-feats.cc:170) Number of frames read 0 was less than target number 400000, using all we read.
ASSERTION_FAILED (gmm-global-init-from-feats[5.5.839~8-0c6a]:Resize():diag-gmm.cc:67) Assertion failed: (nmix > 0 && dim > 0)

[ Stack-Trace: ]
/home/server/kaldi-trunk/src/lib/libkaldi-base.so(kaldi::MessageLogger::LogMessage() const+0xb42) [0x7fe63bed7692]
/home/server/kaldi-trunk/src/lib/libkaldi-base.so(kaldi::KaldiAssertFailure_(char const*, char const*, int, char const*)+0x6e) [0x7fe63bed838e]
/home/server/kaldi-trunk/src/lib/libkaldi-gmm.so(kaldi::DiagGmm::Resize(int, int)+0xb9) [0x7fe63c5d01c3]
gmm-global-init-from-feats(main+0xbd9) [0x55fdf706686f]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xe7) [0x7fe63b341bf7]
gmm-global-init-from-feats(_start+0x2a) [0x55fdf706505a]

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)内劳动服务器公司办公楼一层