add

.gitattributes

@@ -25,3 +25,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+exp/ filter=lfs diff=lfs merge=lfs -text

data/en_token_list/bpe_unigram500/bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1715c2ba8f90e3f660424685a88940b31c31d31acd896dbae677a089e4db9154
+size 245040

exp/asr_train_asr_streaming_transformer_raw_en_bpe500_sp/RESULTS.md

@@ -0,0 +1,35 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Thu Nov 11 09:45:45 CST 2021`
+- python version: `3.7.11 (default, Jul 27 2021, 14:32:16)  [GCC 7.5.0]`
+- espnet version: `espnet 0.9.8`
+- pytorch version: `pytorch 1.5.1`
+- Git hash: `456e6517a47ef71d1b569cfa38b107538d9ef581`
+  - Commit date: `Thu Aug 19 00:48:13 2021 +0800`
+
+## asr_train_asr_streaming_transformer_raw_en_bpe500_sp
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.ave/dev|466|14671|90.3|7.5|2.3|1.7|11.4|82.4|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.ave/test|1155|27500|90.6|6.7|2.8|1.4|10.8|77.5|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.best/test|1155|27500|89.1|7.6|3.3|1.6|12.4|80.8|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.ave/dev|466|78259|96.1|1.6|2.3|1.5|5.4|82.4|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.ave/test|1155|145066|95.9|1.5|2.6|1.2|5.3|77.5|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.best/test|1155|145066|95.1|1.7|3.2|1.4|6.3|80.8|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.ave/dev|466|27927|92.2|5.1|2.7|1.8|9.6|82.4|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.ave/test|1155|51430|92.4|4.6|3.0|1.3|8.9|77.5|
+|decode_asr_streaming_lm_lm_train_lm_en_bpe500_valid.loss.ave_asr_model_valid.acc.best/test|1155|51430|91.0|5.3|3.7|1.5|10.5|80.8|
+

exp/asr_train_asr_streaming_transformer_raw_en_bpe500_sp/config.yaml

@@ -0,0 +1,680 @@
+config: conf/train_asr_streaming_transformer.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/asr_train_asr_streaming_transformer_raw_en_bpe500_sp
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 0
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 4
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 49721
+dist_launcher: null
+multiprocessing_distributed: true
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 20
+patience: 3
+val_scheduler_criterion:
+- valid
+- acc
+early_stopping_criterion:
+- valid
+- acc
+- max
+best_model_criterion:
+-   - valid
+    - acc
+    - max
+keep_nbest_models: 10
+grad_clip: 5
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 128
+valid_batch_size: null
+batch_bins: 1000000
+valid_batch_bins: null
+train_shape_file:
+- exp/asr_stats_raw_en_bpe500_sp/train/speech_shape
+- exp/asr_stats_raw_en_bpe500_sp/train/text_shape.bpe
+valid_shape_file:
+- exp/asr_stats_raw_en_bpe500_sp/valid/speech_shape
+- exp/asr_stats_raw_en_bpe500_sp/valid/text_shape.bpe
+batch_type: folded
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/train_sp/wav.scp
+    - speech
+    - sound
+-   - dump/raw/train_sp/text
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/wav.scp
+    - speech
+    - sound
+-   - dump/raw/dev/text
+    - text
+    - text
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adam
+optim_conf:
+    lr: 0.001
+scheduler: warmuplr
+scheduler_conf:
+    warmup_steps: 25000
+token_list:
+- <blank>
+- <unk>
+- s
+- ▁the
+- t
+- ▁a
+- e
+- ▁and
+- d
+- ▁to
+- ▁of
+- ''''
+- ing
+- ▁in
+- ▁that
+- ▁i
+- m
+- p
+- a
+- y
+- n
+- i
+- c
+- ▁it
+- ▁we
+- ▁you
+- ed
+- o
+- l
+- ▁is
+- u
+- re
+- ▁
+- g
+- r
+- ▁this
+- ar
+- er
+- b
+- ▁so
+- al
+- in
+- or
+- ▁s
+- h
+- ▁for
+- ▁f
+- ▁be
+- k
+- ▁c
+- ▁do
+- le
+- ▁t
+- ▁re
+- ▁w
+- ve
+- f
+- es
+- it
+- w
+- ▁was
+- ▁e
+- ▁they
+- ic
+- ▁on
+- ly
+- ▁are
+- ra
+- ▁have
+- ▁p
+- ▁what
+- ▁with
+- ▁m
+- ent
+- ▁can
+- 'on'
+- an
+- ▁but
+- en
+- ll
+- v
+- at
+- ri
+- ▁st
+- th
+- ur
+- ▁one
+- ▁he
+- il
+- ce
+- ▁about
+- ▁me
+- ▁de
+- ▁all
+- ▁not
+- ▁g
+- ▁ma
+- ch
+- ▁there
+- ter
+- ▁at
+- ation
+- ▁my
+- ▁as
+- ▁n
+- ▁mo
+- el
+- ▁like
+- ▁people
+- ▁con
+- ▁an
+- ▁or
+- ▁if
+- ro
+- ▁from
+- ▁co
+- ate
+- ▁d
+- ver
+- ▁these
+- ▁now
+- ▁out
+- ▁our
+- ▁know
+- ▁see
+- ▁su
+- ion
+- ▁just
+- ▁ch
+- us
+- ol
+- ▁ex
+- ect
+- ne
+- ▁when
+- ▁o
+- ul
+- ▁b
+- te
+- ▁very
+- ▁go
+- ies
+- ▁think
+- ▁because
+- om
+- ▁who
+- ▁up
+- un
+- tion
+- ▁us
+- ▁le
+- ir
+- ity
+- ▁get
+- ▁pa
+- ▁how
+- as
+- ive
+- ▁by
+- et
+- ▁bo
+- ▁se
+- ▁ca
+- ▁which
+- ▁some
+- ▁no
+- ▁going
+- ▁pro
+- ▁po
+- lo
+- ng
+- ▁look
+- ▁time
+- ▁more
+- ▁had
+- ▁would
+- ▁here
+- am
+- ment
+- ▁really
+- la
+- is
+- ▁your
+- x
+- ▁ho
+- ▁k
+- ge
+- ist
+- ▁them
+- ry
+- ▁two
+- id
+- ck
+- ▁fa
+- ▁world
+- ▁sp
+- ure
+- ▁way
+- ide
+- ▁don
+- ▁want
+- ▁work
+- ▁lo
+- ▁were
+- ru
+- est
+- ers
+- ▁their
+- age
+- ▁li
+- ut
+- ▁ba
+- ▁en
+- ▁other
+- ine
+- ow
+- ▁actually
+- ▁sh
+- able
+- ▁make
+- ▁un
+- ac
+- ▁la
+- op
+- ▁into
+- ▁where
+- ▁could
+- ▁has
+- z
+- vi
+- ous
+- ▁will
+- ▁di
+- ight
+- ▁things
+- ally
+- ▁ro
+- ▁say
+- ▁vi
+- ▁years
+- ▁ra
+- ▁than
+- ▁br
+- ther
+- ical
+- ▁ha
+- ▁comp
+- ▁did
+- ▁then
+- ok
+- ▁new
+- ▁any
+- ound
+- ▁well
+- ▁she
+- um
+- ▁been
+- ▁need
+- ▁right
+- ▁thousand
+- ▁take
+- ▁start
+- ▁something
+- ta
+- ▁over
+- he
+- ▁hundred
+- ▁com
+- ▁kind
+- j
+- ▁first
+- ▁those
+- ▁pre
+- ▁said
+- ize
+- tic
+- ▁thing
+- ▁even
+- ▁fi
+- ia
+- ▁back
+- ▁every
+- ▁little
+- nce
+- ▁car
+- ▁sc
+- ▁much
+- ▁life
+- ▁also
+- ▁hi
+- ▁bi
+- if
+- ▁most
+- os
+- ▁three
+- ▁lot
+- ▁got
+- ▁sa
+- ▁come
+- ▁talk
+- iv
+- ▁te
+- ▁inter
+- ▁per
+- ▁use
+- ant
+- ▁bu
+- ▁part
+- ▁his
+- ▁happen
+- ance
+- ▁mean
+- ish
+- ▁dr
+- ▁day
+- ▁only
+- ▁many
+- ▁different
+- ten
+- ▁show
+- ▁down
+- ▁dis
+- ful
+- ▁good
+- ated
+- ▁idea
+- ▁around
+- ▁ga
+- ▁human
+- ▁five
+- ▁put
+- ▁through
+- ▁why
+- ▁real
+- ▁change
+- ff
+- ▁let
+- ible
+- ▁fact
+- ▁same
+- ▁jo
+- ▁her
+- ▁problem
+- ▁live
+- ▁four
+- ▁year
+- ▁give
+- ▁ri
+- ick
+- ▁though
+- ▁big
+- ▁tell
+- ▁great
+- ▁ta
+- ▁va
+- ition
+- ian
+- ▁system
+- ▁six
+- ▁plan
+- ▁place
+- ▁build
+- ▁called
+- ▁point
+- ▁again
+- ▁twenty
+- ▁percent
+- port
+- ▁ten
+- ▁find
+- ▁after
+- ▁app
+- ▁nine
+- ▁qu
+- ▁fl
+- ▁long
+- ▁eight
+- ▁gene
+- ▁design
+- ▁today
+- ▁made
+- ▁imp
+- ▁should
+- ▁came
+- ▁last
+- ▁learn
+- ▁own
+- ▁turn
+- ▁seven
+- ▁high
+- ▁question
+- ▁person
+- ▁another
+- ▁brain
+- ▁important
+- ▁create
+- ▁act
+- ▁trans
+- ▁hu
+- ▁power
+- ness
+- land
+- ▁play
+- ▁sort
+- ▁old
+- ▁course
+- ▁before
+- ▁feel
+- ▁understand
+- q
+- ▁each
+- ▁million
+- ▁together
+- ▁better
+- ▁ago
+- ▁example
+- ▁help
+- ▁story
+- ▁next
+- ▁hand
+- ▁school
+- ▁develop
+- ▁technology
+- ▁under
+- form
+- ▁second
+- que
+- ▁grow
+- ▁still
+- ▁believe
+- ▁cell
+- ▁light
+- ▁number
+- ▁small
+- ▁between
+- ▁become
+- ▁america
+- ▁maybe
+- ▁space
+- ▁project
+- ▁organ
+- ▁vo
+- ▁home
+- ▁children
+- graph
+- ▁picture
+- ▁fifty
+- ▁africa
+- ▁thirty
+- ▁reason
+- ▁health
+- ▁large
+- ▁hard
+- ▁sense
+- ▁computer
+- ▁always
+- ▁women
+- ▁money
+- ▁everything
+- ▁information
+- ▁country
+- ▁teach
+- ▁energy
+- ▁run
+- ▁experience
+- ▁food
+- ▁process
+- ▁quite
+- qua
+- ▁future
+- ▁interesting
+- ization
+- ▁enough
+- ▁science
+- ▁stuff
+- ▁social
+- ▁probably
+- ▁friend
+- ▁ja
+- ▁half
+- ability
+- ▁group
+- ▁animal
+- ▁wonder
+- ▁product
+- ▁close
+- ▁simple
+- ▁piece
+- ▁billion
+- ▁control
+- ▁sign
+- ▁environment
+- ▁global
+- ▁imagine
+- ▁basically
+- ▁answer
+- ō
+- ă
+- ť
+- <sos/eos>
+init: null
+input_size: null
+ctc_conf:
+    dropout_rate: 0.0
+    ctc_type: builtin
+    reduce: true
+    ignore_nan_grad: false
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1
+    length_normalized_loss: false
+use_preprocessor: true
+token_type: bpe
+bpemodel: data/en_token_list/bpe_unigram500/bpe.model
+non_linguistic_symbols: null
+cleaner: null
+g2p: null
+speech_volume_normalize: null
+rir_scp: null
+rir_apply_prob: 1.0
+noise_scp: null
+noise_apply_prob: 1.0
+noise_db_range: '13_15'
+frontend: default
+frontend_conf:
+    fs: 16k
+specaug: specaug
+specaug_conf:
+    apply_time_warp: true
+    time_warp_window: 5
+    time_warp_mode: bicubic
+    apply_freq_mask: true
+    freq_mask_width_range:
+    - 0
+    - 30
+    num_freq_mask: 2
+    apply_time_mask: true
+    time_mask_width_range:
+    - 0
+    - 40
+    num_time_mask: 2
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/asr_stats_raw_en_bpe500_sp/train/feats_stats.npz
+preencoder: null
+preencoder_conf: {}
+encoder: contextual_block_transformer
+encoder_conf:
+    output_size: 256
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 12
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    attention_dropout_rate: 0.0
+    input_layer: conv2d
+    normalize_before: true
+    block_size: 40
+    hop_size: 16
+    look_ahead: 16
+    init_average: true
+    ctx_pos_enc: true
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 6
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.0
+    src_attention_dropout_rate: 0.0
+required:
+- output_dir
+- token_list
+version: 0.9.8
+distributed: true

exp/asr_train_asr_streaming_transformer_raw_en_bpe500_sp/valid.acc.ave_10best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eef0f947cfd9af0f70343e7200ea8c4db49ad28cf8c5effc3b0026573cbfcf1
+size 110072816

exp/lm_train_lm_en_bpe500/19epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbc4d06746443b60b203e246ab397f9694a3486675f1ffdb74e0df5deec5b003
+size 29685311

exp/lm_train_lm_en_bpe500/config.yaml

@@ -0,0 +1,611 @@
+config: conf/train_lm.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/lm_train_lm_en_bpe500
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: null
+dist_rank: null
+local_rank: 0
+dist_master_addr: null
+dist_master_port: null
+dist_launcher: null
+multiprocessing_distributed: false
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 20
+patience: 3
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+keep_nbest_models: 1
+grad_clip: 5.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 64
+valid_batch_size: null
+batch_bins: 1000000
+valid_batch_bins: null
+train_shape_file:
+- exp/lm_stats_en_bpe500/train/text_shape.bpe
+valid_shape_file:
+- exp/lm_stats_en_bpe500/valid/text_shape.bpe
+batch_type: folded
+valid_batch_type: null
+fold_length:
+- 150
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/lm_train.txt
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/text
+    - text
+    - text
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: sgd
+optim_conf: {}
+scheduler: null
+scheduler_conf: {}
+token_list:
+- <blank>
+- <unk>
+- s
+- ▁the
+- t
+- ▁a
+- e
+- ▁and
+- d
+- ▁to
+- ▁of
+- ''''
+- ing
+- ▁in
+- ▁that
+- ▁i
+- m
+- p
+- a
+- y
+- n
+- i
+- c
+- ▁it
+- ▁we
+- ▁you
+- ed
+- o
+- l
+- ▁is
+- u
+- re
+- ▁
+- g
+- r
+- ▁this
+- ar
+- er
+- b
+- ▁so
+- al
+- in
+- or
+- ▁s
+- h
+- ▁for
+- ▁f
+- ▁be
+- k
+- ▁c
+- ▁do
+- le
+- ▁t
+- ▁re
+- ▁w
+- ve
+- f
+- es
+- it
+- w
+- ▁was
+- ▁e
+- ▁they
+- ic
+- ▁on
+- ly
+- ▁are
+- ra
+- ▁have
+- ▁p
+- ▁what
+- ▁with
+- ▁m
+- ent
+- ▁can
+- 'on'
+- an
+- ▁but
+- en
+- ll
+- v
+- at
+- ri
+- ▁st
+- th
+- ur
+- ▁one
+- ▁he
+- il
+- ce
+- ▁about
+- ▁me
+- ▁de
+- ▁all
+- ▁not
+- ▁g
+- ▁ma
+- ch
+- ▁there
+- ter
+- ▁at
+- ation
+- ▁my
+- ▁as
+- ▁n
+- ▁mo
+- el
+- ▁like
+- ▁people
+- ▁con
+- ▁an
+- ▁or
+- ▁if
+- ro
+- ▁from
+- ▁co
+- ate
+- ▁d
+- ver
+- ▁these
+- ▁now
+- ▁out
+- ▁our
+- ▁know
+- ▁see
+- ▁su
+- ion
+- ▁just
+- ▁ch
+- us
+- ol
+- ▁ex
+- ect
+- ne
+- ▁when
+- ▁o
+- ul
+- ▁b
+- te
+- ▁very
+- ▁go
+- ies
+- ▁think
+- ▁because
+- om
+- ▁who
+- ▁up
+- un
+- tion
+- ▁us
+- ▁le
+- ir
+- ity
+- ▁get
+- ▁pa
+- ▁how
+- as
+- ive
+- ▁by
+- et
+- ▁bo
+- ▁se
+- ▁ca
+- ▁which
+- ▁some
+- ▁no
+- ▁going
+- ▁pro
+- ▁po
+- lo
+- ng
+- ▁look
+- ▁time
+- ▁more
+- ▁had
+- ▁would
+- ▁here
+- am
+- ment
+- ▁really
+- la
+- is
+- ▁your
+- x
+- ▁ho
+- ▁k
+- ge
+- ist
+- ▁them
+- ry
+- ▁two
+- id
+- ck
+- ▁fa
+- ▁world
+- ▁sp
+- ure
+- ▁way
+- ide
+- ▁don
+- ▁want
+- ▁work
+- ▁lo
+- ▁were
+- ru
+- est
+- ers
+- ▁their
+- age
+- ▁li
+- ut
+- ▁ba
+- ▁en
+- ▁other
+- ine
+- ow
+- ▁actually
+- ▁sh
+- able
+- ▁make
+- ▁un
+- ac
+- ▁la
+- op
+- ▁into
+- ▁where
+- ▁could
+- ▁has
+- z
+- vi
+- ous
+- ▁will
+- ▁di
+- ight
+- ▁things
+- ally
+- ▁ro
+- ▁say
+- ▁vi
+- ▁years
+- ▁ra
+- ▁than
+- ▁br
+- ther
+- ical
+- ▁ha
+- ▁comp
+- ▁did
+- ▁then
+- ok
+- ▁new
+- ▁any
+- ound
+- ▁well
+- ▁she
+- um
+- ▁been
+- ▁need
+- ▁right
+- ▁thousand
+- ▁take
+- ▁start
+- ▁something
+- ta
+- ▁over
+- he
+- ▁hundred
+- ▁com
+- ▁kind
+- j
+- ▁first
+- ▁those
+- ▁pre
+- ▁said
+- ize
+- tic
+- ▁thing
+- ▁even
+- ▁fi
+- ia
+- ▁back
+- ▁every
+- ▁little
+- nce
+- ▁car
+- ▁sc
+- ▁much
+- ▁life
+- ▁also
+- ▁hi
+- ▁bi
+- if
+- ▁most
+- os
+- ▁three
+- ▁lot
+- ▁got
+- ▁sa
+- ▁come
+- ▁talk
+- iv
+- ▁te
+- ▁inter
+- ▁per
+- ▁use
+- ant
+- ▁bu
+- ▁part
+- ▁his
+- ▁happen
+- ance
+- ▁mean
+- ish
+- ▁dr
+- ▁day
+- ▁only
+- ▁many
+- ▁different
+- ten
+- ▁show
+- ▁down
+- ▁dis
+- ful
+- ▁good
+- ated
+- ▁idea
+- ▁around
+- ▁ga
+- ▁human
+- ▁five
+- ▁put
+- ▁through
+- ▁why
+- ▁real
+- ▁change
+- ff
+- ▁let
+- ible
+- ▁fact
+- ▁same
+- ▁jo
+- ▁her
+- ▁problem
+- ▁live
+- ▁four
+- ▁year
+- ▁give
+- ▁ri
+- ick
+- ▁though
+- ▁big
+- ▁tell
+- ▁great
+- ▁ta
+- ▁va
+- ition
+- ian
+- ▁system
+- ▁six
+- ▁plan
+- ▁place
+- ▁build
+- ▁called
+- ▁point
+- ▁again
+- ▁twenty
+- ▁percent
+- port
+- ▁ten
+- ▁find
+- ▁after
+- ▁app
+- ▁nine
+- ▁qu
+- ▁fl
+- ▁long
+- ▁eight
+- ▁gene
+- ▁design
+- ▁today
+- ▁made
+- ▁imp
+- ▁should
+- ▁came
+- ▁last
+- ▁learn
+- ▁own
+- ▁turn
+- ▁seven
+- ▁high
+- ▁question
+- ▁person
+- ▁another
+- ▁brain
+- ▁important
+- ▁create
+- ▁act
+- ▁trans
+- ▁hu
+- ▁power
+- ness
+- land
+- ▁play
+- ▁sort
+- ▁old
+- ▁course
+- ▁before
+- ▁feel
+- ▁understand
+- q
+- ▁each
+- ▁million
+- ▁together
+- ▁better
+- ▁ago
+- ▁example
+- ▁help
+- ▁story
+- ▁next
+- ▁hand
+- ▁school
+- ▁develop
+- ▁technology
+- ▁under
+- form
+- ▁second
+- que
+- ▁grow
+- ▁still
+- ▁believe
+- ▁cell
+- ▁light
+- ▁number
+- ▁small
+- ▁between
+- ▁become
+- ▁america
+- ▁maybe
+- ▁space
+- ▁project
+- ▁organ
+- ▁vo
+- ▁home
+- ▁children
+- graph
+- ▁picture
+- ▁fifty
+- ▁africa
+- ▁thirty
+- ▁reason
+- ▁health
+- ▁large
+- ▁hard
+- ▁sense
+- ▁computer
+- ▁always
+- ▁women
+- ▁money
+- ▁everything
+- ▁information
+- ▁country
+- ▁teach
+- ▁energy
+- ▁run
+- ▁experience
+- ▁food
+- ▁process
+- ▁quite
+- qua
+- ▁future
+- ▁interesting
+- ization
+- ▁enough
+- ▁science
+- ▁stuff
+- ▁social
+- ▁probably
+- ▁friend
+- ▁ja
+- ▁half
+- ability
+- ▁group
+- ▁animal
+- ▁wonder
+- ▁product
+- ▁close
+- ▁simple
+- ▁piece
+- ▁billion
+- ▁control
+- ▁sign
+- ▁environment
+- ▁global
+- ▁imagine
+- ▁basically
+- ▁answer
+- ō
+- ă
+- ť
+- <sos/eos>
+init: null
+model_conf:
+    ignore_id: 0
+use_preprocessor: true
+token_type: bpe
+bpemodel: data/en_token_list/bpe_unigram500/bpe.model
+non_linguistic_symbols: null
+cleaner: null
+g2p: null
+lm: seq_rnn
+lm_conf:
+    unit: 650
+    nlayers: 2
+required:
+- output_dir
+- token_list
+version: 0.9.8
+distributed: false

exp/lm_train_lm_en_bpe500/perplexity_test/ppl

@@ -0,0 +1 @@
+22.910754869384046

meta.yaml

@@ -0,0 +1,10 @@
+espnet: 0.9.8
+files:
+  asr_model_file: exp/asr_train_asr_streaming_transformer_raw_en_bpe500_sp/valid.acc.ave_10best.pth
+  lm_file: exp/lm_train_lm_en_bpe500/19epoch.pth
+python: "3.7.11 (default, Jul 27 2021, 14:32:16) \n[GCC 7.5.0]"
+timestamp: 1636877168.750594
+torch: 1.5.1
+yaml_files:
+  asr_train_config: exp/asr_train_asr_streaming_transformer_raw_en_bpe500_sp/config.yaml
+  lm_train_config: exp/lm_train_lm_en_bpe500/config.yaml