pytorch-kaldi-neural-speaker-embeddings
A light weight neural speaker embeddings extraction based on Kaldi and PyTorch.
파일 탐색기
최종 버전 다운로드 (.zip)- fbank.conf
- mfcc.conf
- slurm.conf
- vad.conf
- ivector_43873_tsne_2d_plot.png
- ivector_meannorm_200_lda_lennorm_43873_tsne_2d_plot.png
- meannorm_200_lda_lennorm_resnet_mfcc_3-8_200_32_mean_lde_sqr_asoftmax_m2_43873_tsne_2d_plot.png
- meannorm_200_lda_lennorm_resnet_mfcc_3-8_256_32_mean_lde_sqr_asoftmax_m2_43873_tsne_2d_plot.png
- meannorm_200_lda_lennorm_resnet_mfcc_3-8_512_32_mean+std_lde_sqr_asoftmax_m2_43873_tsne_2d_plot.png
- meannorm_200_lda_lennorm_resnet_mfcc_3-8_512_32_mean_lde_sqr_asoftmax_m2_43873_tsne_2d_plot.png
- meannorm_200_lda_lennorm_resnet_mfcc_3-8_512_32_mean_lde_sqr_asoftmax_m3_43873_tsne_2d_plot.png
- meannorm_200_lda_lennorm_resnet_mfcc_3-8_512_32_mean_lde_sqr_asoftmax_m4_43873_tsne_2d_plot.png
- meannorm_200_lda_lennorm_resnet_mfcc_3-8_512_32_mean_lde_sqr_softmax_43873_tsne_2d_plot.png
- resnet_mfcc_3-8_200_32_mean_lde_sqr_asoftmax_m2_43873_tsne_2d_plot.png
- resnet_mfcc_3-8_256_32_mean_lde_sqr_asoftmax_m2_43873_tsne_2d_plot.png
- resnet_mfcc_3-8_512_32_mean+std_lde_sqr_asoftmax_m2_43873_tsne_2d_plot.png
- resnet_mfcc_3-8_512_32_mean_lde_sqr_asoftmax_m2_43873_tsne_2d_plot.png
- resnet_mfcc_3-8_512_32_mean_lde_sqr_asoftmax_m3_43873_tsne_2d_plot.png
- resnet_mfcc_3-8_512_32_mean_lde_sqr_asoftmax_m4_43873_tsne_2d_plot.png
- resnet_mfcc_3-8_512_32_mean_lde_sqr_softmax_43873_tsne_2d_plot.png
- .DS_Store
- LDE-6.pdf
- LDE-6.png
- TTS-4.pdf
- TTS-4.png
- run_xvector_1a.sh
- prepare_feats_for_egs.sh
- run_xvector.sh
- add_disambig.pl
- add_lex_disambig.pl
- analyze_segments.pl
- apply_map.pl
- best_wer.sh
- build_const_arpa_lm.sh
- check_spk_emb_range.py
- combine_data.sh
- compute_min_dcf.py
- compute_vad_decision.sh
- convert_ctm.pl
- convert_slf.pl
- convert_slf_parallel.sh
- copy_data_dir.sh
- create_data_link.pl
- create_split_dir.pl
- dict_dir_add_pronprobs.sh
- eps2disambig.pl
- filt.py
- filter_scp.pl
- filter_scps.pl
- find_arpa_oovs.pl
- fix_ctm.sh
- fix_data_dir.sh
- format_lm.sh
- format_lm_sri.sh
- gen_topo.pl
- generate_vctk_wav.py
- get_spk_emb.py
- get_spk_emb_2.py
- get_utt2num_frames.sh
- int2sym.pl
- kwslist_post_process.pl
- ln.pl
- make_absolute.sh
- make_fbank.sh
- make_lexicon_fst.pl
- make_lexicon_fst_silprob.pl
- make_mfcc.sh
- make_musan.py
- make_musan.sh
- make_unigram_grammar.pl
- make_vctk.pl
- make_vctk_wav.py
- make_vctk_wav.sh
- make_voxceleb1.pl
- make_voxceleb2.pl
- map_arpa_lm.pl
- mkgraph.sh
- parse_options.sh
- pbs.pl
- perturb_data_dir_speed.sh
- pinyin_map.pl
- prepare_extended_lang.sh
- prepare_for_eer.py
- prepare_lang.sh
- prepare_online_nnet_dist_build.sh
- queue.pl
- remove_data_links.sh
- remove_oovs.pl
- retry.pl
- reverse_arpa.py
- rnnlm_compute_scores.sh
- run.pl
- s2eps.pl
- segmentation.pl
- show_lattice.sh
- shuffle_list.pl
- slurm.pl
- spk2utt_to_utt2spk.pl
- split_data.sh
- split_scp.pl
- ssh.pl
- subset_data_dir.sh
- subset_data_dir_tr_cv.sh
- subset_scp.pl
- summarize_logs.pl
- summarize_warnings.pl
- sym2int.pl
- utt2spk_to_spk2utt.pl
- validate_data_dir.sh
- validate_dict_dir.pl
- validate_lang.pl
- validate_text.pl
- visualize_spk_emb.py
- visualize_trait_emb.py
- visualize_utt_emb.py
- write_kwslist.pl
- datasets.cpython-37.pyc
- densenet.cpython-35.pyc
- densenet.cpython-36.pyc
- densenet.cpython-37.pyc
- kaldi_io.cpython-35.pyc
- kaldi_io.cpython-36.pyc
- kaldi_io.cpython-37.pyc
- model.cpython-35.pyc
- model.cpython-36.pyc
- model.cpython-37.pyc
- model2.cpython-35.pyc
- model3.cpython-35.pyc
- model4.cpython-35.pyc
- SequenceDataset5.cpython-35.pyc
- SequenceDataset5.cpython-36.pyc
- datasets.py
- decode.py
- densenet.py
- kaldi_io.py
- main.py
- model.py
- prepare_data.py
- cmd.sh
- LICENSE
- path.sh
- pipeline.sh
- README.md
- requirements.txt
// repository documentation
Was this content helpful?
(0 ratings)
