1

我正在使用 gooofy zamia-speech 为 kaldi 的模型适应项目。

我按照kaldi-adapt-lm给出的步骤使用kaldi-generic-de-tdnn_f-r20190328 model 创建模型

当我在 .wav 文件上对其进行测试时,它显示以下错误:


//////
nnet3-latgen-faster --frame-subsampling-factor=3 --frames-per-chunk=50 --extra-left-context=0 --extra-right-context=0 --extra-left-context-initial=-1 --extra-right-context-final=-1 --minimize=false --max-active=7000 --min-active=200 --beam=15.0 --lattice-beam=8.0 --acoustic-scale=1.0 --allow-partial=true --word-symbol-table=exp/api.ai-model/words.txt exp/api.ai-model/final.mdl exp/api.ai-model//HCLG.fst 'ark,s,cs:apply-cmvn --norm-means=false --norm-vars=false --utt2spk=ark:data/test-corpus/utt2spk scp:data/test-corpus/cmvn.scp scp:data/test-corpus/feats.scp ark:- |' 'ark:|lattice-scale --acoustic-scale=10.0 ark:- ark:-  >exp/lat.1' 
LOG (nnet3-latgen-faster[5.5.628~1-ac6b1]:RemoveOrphanNodes():nnet-nnet.cc:948) Removed 1 orphan nodes.
LOG (nnet3-latgen-faster[5.5.628~1-ac6b1]:RemoveOrphanComponents():nnet-nnet.cc:847) Removing 2 orphan components.
LOG (nnet3-latgen-faster[5.5.628~1-ac6b1]:Collapse():nnet-utils.cc:1472) Added 1 components, removed 2
lattice-scale --acoustic-scale=10.0 ark:- ark:- 
apply-cmvn --norm-means=false --norm-vars=false --utt2spk=ark:data/test-corpus/utt2spk scp:data/test-corpus/cmvn.scp scp:data/test-corpus/feats.scp ark:- 
LOG (apply-cmvn[5.5.628~1-ac6b1]:main():apply-cmvn.cc:81) Copied 1 utterances.
LOG (nnet3-latgen-faster[5.5.628~1-ac6b1]:CheckAndFixConfigs():nnet-am-decodable-simple.cc:294) Increasing --frames-per-chunk from 50 to 51 to make it a multiple of --frame-subsampling-factor=3
ERROR (nnet3-latgen-faster[5.5.628~1-ac6b1]:EnsureFrameIsComputed():nnet-am-decodable-simple.cc:105) Neural net expects 'ivector' features with dimension 100 but you provided 0

[ Stack-Trace: ]
nnet3-latgen-faster(kaldi::MessageLogger::LogMessage() const+0x82c) [0x8ab31c]
nnet3-latgen-faster(kaldi::MessageLogger::LogAndThrow::operator=(kaldi::MessageLogger const&)+0x21) [0x55786b]
nnet3-latgen-faster(kaldi::nnet3::DecodableNnetSimple::EnsureFrameIsComputed(int)+0x298) [0x5dbcfa]
nnet3-latgen-faster(kaldi::nnet3::DecodableAmNnetSimple::LogLikelihood(int, int)+0x40) [0x5dc27a]
nnet3-latgen-faster(kaldi::LatticeFasterDecoderTpl<fst::ConstFst<fst::ArcTpl<fst::TropicalWeightTpl<float> >, unsigned int>, kaldi::decoder::StdToken>::ProcessEmitting(kaldi::DecodableInterface*)+0x22f) [0x73873d]
nnet3-latgen-faster(kaldi::LatticeFasterDecoderTpl<fst::ConstFst<fst::ArcTpl<fst::TropicalWeightTpl<float> >, unsigned int>, kaldi::decoder::StdToken>::AdvanceDecoding(kaldi::DecodableInterface*, int)+0x97) [0x738b1d]
nnet3-latgen-faster(kaldi::LatticeFasterDecoderTpl<fst::Fst<fst::ArcTpl<fst::TropicalWeightTpl<float> > >, kaldi::decoder::StdToken>::AdvanceDecoding(kaldi::DecodableInterface*, int)+0x41) [0x738bc9]
nnet3-latgen-faster(kaldi::LatticeFasterDecoderTpl<fst::Fst<fst::ArcTpl<fst::TropicalWeightTpl<float> > >, kaldi::decoder::StdToken>::Decode(kaldi::DecodableInterface*)+0x21) [0x738d19]
nnet3-latgen-faster(bool kaldi::DecodeUtteranceLatticeFaster<fst::Fst<fst::ArcTpl<fst::TropicalWeightTpl<float> > > >(kaldi::LatticeFasterDecoderTpl<fst::Fst<fst::ArcTpl<fst::TropicalWeightTpl<float> > >, kaldi::decoder::StdToken>&, kaldi::DecodableInterface&, kaldi::TransitionModel const&, fst::SymbolTable const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, double, bool, bool, kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, kaldi::TableWriter<kaldi::CompactLatticeHolder>*, kaldi::TableWriter<kaldi::LatticeHolder>*, double*)+0x88) [0x76a629]
nnet3-latgen-faster(main+0xfd2) [0x553c98]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf0) [0x7fbe85da5840]
nnet3-latgen-faster(_start+0x29) [0x552bf9]

WARNING (nnet3-latgen-faster[5.5.628~1-ac6b1]:~HashList():util/hash-list-inl.h:117) Possible memory leak: 1023 != 1024: you might have forgotten to call Delete on some Elems
LOG (lattice-scale[5.5.628~1-ac6b1]:main():lattice-scale.cc:107) Done 0 lattices.
WARNING (nnet3-latgen-faster[5.5.628~1-ac6b1]:Close():kaldi-io.cc:333) Pipe |lattice-scale --acoustic-scale=10.0 ark:- ark:-  >exp/lat.1 had nonzero return status 256

有人可以帮我解决吗?

先感谢您!

4

0 回答 0