icefall-asr-librispeech-pruned-stateless-streaming-conformer-rnnt4-2022-06-10
/
decoding_results
/greedy_search
/log-decode-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model-2022-05-30-11-41-03
2022-05-30 11:41:03,602 INFO [decode.py:595] Decoding started | |
2022-05-30 11:41:03,602 INFO [decode.py:601] Device: cuda:0 | |
2022-05-30 11:41:03,604 INFO [decode.py:616] {'best_train_loss': inf, 'best_valid_loss': inf, 'best_train_epoch': -1, 'best_valid_epoch': -1, 'batch_idx_train': 0, 'log_interval': 50, 'reset_interval': 200, 'valid_interval': 3000, 'feature_dim': 80, 'subsampling_factor': 4, 'encoder_dim': 512, 'nhead': 8, 'dim_feedforward': 2048, 'num_encoder_layers': 12, 'decoder_dim': 512, 'joiner_dim': 512, 'model_warm_step': 3000, 'env_info': {'k2-version': '1.15.1', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': 'f8d2dba06c000ffee36aab5b66f24e7c9809f116', 'k2-git-date': 'Thu Apr 21 12:20:34 2022', 'lhotse-version': '1.1.0', 'torch-version': '1.10.0+cu102', 'torch-cuda-available': True, 'torch-cuda-version': '10.2', 'python-version': '3.8', 'icefall-git-branch': 'streaming-conformer', 'icefall-git-sha1': '0325e3a-dirty', 'icefall-git-date': 'Sun May 29 16:25:01 2022', 'icefall-path': '/ceph-kw/kangwei/code/icefall_reworked2', 'k2-path': '/ceph-hw/kangwei/code/k2_release/k2/k2/python/k2/__init__.py', 'lhotse-path': '/ceph-hw/kangwei/dev_tools/anaconda3/envs/rnnt2/lib/python3.8/site-packages/lhotse-1.1.0-py3.8.egg/lhotse/__init__.py', 'hostname': 'de-74279-k2-train-2-0307200233-b554c565c-lf9qd', 'IP address': '10.177.74.201'}, 'epoch': 29, 'iter': 0, 'avg': 6, 'use_averaged_model': True, 'exp_dir': PosixPath('streaming_pruned_transducer_stateless4/exp'), 'bpe_model': 'data/lang_bpe_500/bpe.model', 'decoding_method': 'greedy_search', 'beam_size': 4, 'beam': 4, 'max_contexts': 4, 'max_states': 8, 'context_size': 2, 'max_sym_per_frame': 1, 'dynamic_chunk_training': False, 'short_chunk_size': 25, 'num_left_chunks': 4, 'simulate_streaming': True, 'causal_convolution': True, 'right_chunk_size': 16, 'left_context': 64, 'full_libri': True, 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 1000, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 2, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'res_dir': PosixPath('streaming_pruned_transducer_stateless4/exp/greedy_search'), 'suffix': 'epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model', 'blank_id': 0, 'unk_id': 2, 'vocab_size': 500} | |
2022-05-30 11:41:03,604 INFO [decode.py:618] About to create model | |
2022-05-30 11:41:04,044 INFO [decode.py:685] Calculating the averaged model over epoch range from 23 (excluded) to 29 | |
2022-05-30 11:41:11,215 INFO [decode.py:707] Number of model parameters: 78648040 | |
2022-05-30 11:41:11,215 INFO [asr_datamodule.py:422] About to get test-clean cuts | |
2022-05-30 11:41:11,440 INFO [asr_datamodule.py:427] About to get test-other cuts | |
2022-05-30 11:41:12,944 INFO [decode.py:502] batch 0/?, cuts processed until now is 205 | |
2022-05-30 11:41:30,041 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt | |
2022-05-30 11:41:30,134 INFO [utils.py:405] [test-clean-greedy_search] %WER 3.35% [1763 / 52576, 182 ins, 158 del, 1423 sub ] | |
2022-05-30 11:41:30,381 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-clean-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt | |
2022-05-30 11:41:30,382 INFO [decode.py:549] | |
For test-clean, WER of different settings are: | |
greedy_search 3.35 best for test-clean | |
2022-05-30 11:41:31,731 INFO [decode.py:502] batch 0/?, cuts processed until now is 231 | |
2022-05-30 11:41:47,728 INFO [decode.py:519] The transcripts are stored in streaming_pruned_transducer_stateless4/exp/greedy_search/recogs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt | |
2022-05-30 11:41:47,803 INFO [utils.py:405] [test-other-greedy_search] %WER 8.62% [4514 / 52343, 456 ins, 465 del, 3593 sub ] | |
2022-05-30 11:41:48,030 INFO [decode.py:532] Wrote detailed error stats to streaming_pruned_transducer_stateless4/exp/greedy_search/errs-test-other-greedy_search-epoch-29-avg-6-streaming-chunk-size-16-left-context-64-context-2-max-sym-per-frame-1-use-averaged-model.txt | |
2022-05-30 11:41:48,031 INFO [decode.py:549] | |
For test-other, WER of different settings are: | |
greedy_search 8.62 best for test-other | |
2022-05-30 11:41:48,031 INFO [decode.py:735] Done! | |