lalimili6

My idea that GPU has more computation capacity and can decoding more waves in realtime. TCP decoding can decode only one wave on one TCP port. A server has about...

Here is log file: [steps/nnet3/decode_semisup.sh](https://github.com/kaldi-asr/kaldi/blob/master/egs/wsj/s5/steps/nnet3/decode_semisup.sh#L111) ``` # nnet3-latgen-faster --online-ivectors=scp:exp/nnet3/ivectors_test_hires//ivector_online.scp --online-ivector-period=10 --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 --word-determinize=false --max-active=7000 --min-active=200 --beam=15.0 --lattice-beam=8.0 --acoustic-scale=1.0 --allow-partial=true --word-symbol-table=model/graph_test//words.txt --determinize-lattice=false model_online/final.mdl model/graph_test//HCLG.fst "ark,s,cs:apply-cmvn --utt2spk=ark:data/test_hires/split4/1/utt2spk...

> You apply CMN in nnet3/decode_semisup.sh but not in gpu_semisup.sh (variance normalization if off, mean on by default in `apply-cmvn`). Is this the root cause? GPU decoding (batched-wav-nnet3-cuda2 with or...

Many thanks. Yes I did that and add TopSortLatticeIfNeeded() to [latbin/lattice-determinize-phone-pruned.cc line 104](https://github.com/kaldi-asr/kaldi/blob/master/src/latbin/lattice-determinize-phone-pruned.cc#L104). is it true? but results like my first post.

Here are results https://github.com/k2-fsa/sherpa/pull/168 many thanks

@nshmyrev I think this is a good and new example. what your opinion? https://github.com/kaldi-asr/kaldi/blob/master/egs/libri_css/s5_mono/local/chain/tuning/run_tdnn_1d_ft.sh