Simpleoier Librispeech Asr Train Asr Conformer7 Wavlm Large Raw En Bpe5000 Sp
S
Simpleoier Librispeech Asr Train Asr Conformer7 Wavlm Large Raw En Bpe5000 Sp
由espnet開發
基於ESPnet框架訓練的自動語音識別(ASR)模型,使用Conformer架構和WavLM大型預訓練模型,在LibriSpeech數據集上訓練。
下載量 66
發布時間 : 3/2/2022
模型概述
該模型是一個高性能的英語自動語音識別系統,專為處理原始音頻輸入並轉換為文本而設計。
模型特點
高性能架構
結合Conformer7和WavLM大型預訓練模型,提供卓越的語音識別能力
LibriSpeech訓練
在廣泛使用的LibriSpeech數據集上訓練,確保模型在多種語音條件下的魯棒性
低錯誤率
在測試集上表現出色,詞錯誤率(WER)在乾淨語音上低至1.8%,在嘈雜語音上為3.7%
模型能力
英語語音識別
原始音頻處理
大規模語音轉文本
使用案例
語音轉錄
會議記錄
自動轉錄會議錄音
準確率高達98.4%(測試集clean數據)
音頻字幕生成
為播客或視頻內容生成字幕
在嘈雜語音環境下仍保持96.7%準確率
語音助手
語音命令識別
識別和執行語音命令
🚀 ESPnet2 ASR模型
本模型是基於ESPnet的自動語音識別(ASR)模型,利用LibriSpeech數據集進行訓練,能夠實現高效準確的語音識別功能。
🚀 快速開始
環境準備
該模型由simpleoier使用espnet中的librispeech配方進行訓練。
運行示例
以下是在ESPnet2中使用該模型的示例:
cd espnet
git checkout b0ff60946ada6753af79423a2e6063984bec2926
pip install -e .
cd egs2/librispeech/asr1
./run.sh --skip_data_prep false --skip_train true --download_model espnet/simpleoier_librispeech_asr_train_asr_conformer7_wavlm_large_raw_en_bpe5000_sp
📚 詳細文檔
實驗結果
環境信息
- 日期:
Tue Jan 4 20:52:48 EST 2022
- Python版本:
3.7.11 (default, Jul 27 2021, 14:32:16) [GCC 7.5.0]
- ESPnet版本:
espnet 0.10.5a1
- PyTorch版本:
pytorch 1.8.1
- Git哈希值: ``
- 提交日期: ``
WER(詞錯誤率)
數據集 | 句子數 | 詞數 | 正確詞數 | 替換詞數 | 刪除詞數 | 插入詞數 | 錯誤率 | 句子錯誤率 |
---|---|---|---|---|---|---|---|---|
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/dev_clean | 2703 | 54402 | 98.4 | 1.4 | 0.1 | 0.2 | 1.7 | 23.1 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/dev_other | 2864 | 50948 | 96.7 | 3.0 | 0.3 | 0.3 | 3.6 | 35.5 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/test_clean | 2620 | 52576 | 98.4 | 1.5 | 0.1 | 0.2 | 1.8 | 23.7 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/test_other | 2939 | 52343 | 96.7 | 3.0 | 0.3 | 0.4 | 3.7 | 37.9 |
CER(字符錯誤率)
數據集 | 句子數 | 字符數 | 正確字符數 | 替換字符數 | 刪除字符數 | 插入字符數 | 錯誤率 | 句子錯誤率 |
---|---|---|---|---|---|---|---|---|
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/dev_clean | 2703 | 288456 | 99.7 | 0.2 | 0.2 | 0.2 | 0.5 | 23.1 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/dev_other | 2864 | 265951 | 98.9 | 0.6 | 0.4 | 0.4 | 1.5 | 35.5 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/test_clean | 2620 | 281530 | 99.6 | 0.2 | 0.2 | 0.2 | 0.6 | 23.7 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/test_other | 2939 | 272758 | 99.1 | 0.5 | 0.4 | 0.4 | 1.3 | 37.9 |
TER(詞塊錯誤率)
數據集 | 句子數 | 詞塊數 | 正確詞塊數 | 替換詞塊數 | 刪除詞塊數 | 插入詞塊數 | 錯誤率 | 句子錯誤率 |
---|---|---|---|---|---|---|---|---|
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/dev_clean | 2703 | 68010 | 98.2 | 1.4 | 0.4 | 0.3 | 2.1 | 23.1 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/dev_other | 2864 | 63110 | 96.0 | 3.1 | 0.9 | 0.9 | 4.9 | 35.5 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/test_clean | 2620 | 65818 | 98.1 | 1.4 | 0.5 | 0.4 | 2.3 | 23.7 |
decode_asr_lm_lm_train_lm_transformer2_en_bpe5000_valid.loss.ave_asr_model_valid.acc.ave/test_other | 2939 | 65101 | 96.1 | 2.9 | 1.0 | 0.8 | 4.7 | 37.9 |
ASR配置
展開查看
config: conf/tuning/train_asr_conformer7_wavlm_large.yaml
print_config: false
log_level: INFO
dry_run: false
iterator_type: sequence
output_dir: exp/asr_train_asr_conformer7_wavlm_large_raw_en_bpe5000_sp
ngpu: 1
seed: 0
num_workers: 1
num_att_plot: 3
num_targets: 1
dist_backend: nccl
dist_init_method: env://
dist_world_size: 2
dist_rank: 0
local_rank: 0
dist_master_addr: localhost
dist_master_port: 45342
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: 35
patience: null
val_scheduler_criterion:
- valid
- loss
early_stopping_criterion:
- valid
- loss
- min
best_model_criterion:
- - valid
- acc
- max
keep_nbest_models: 10
nbest_averaging_interval: 0
grad_clip: 5.0
grad_clip_type: 2.0
grad_noise: false
accum_grad: 3
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
wandb_entity: null
wandb_name: null
wandb_model_log_interval: -1
detect_anomaly: false
pretrain_path: null
init_param: []
ignore_init_mismatch: false
freeze_param:
- frontend.upstream
num_iters_per_epoch: null
batch_size: 20
valid_batch_size: null
batch_bins: 40000000
valid_batch_bins: null
train_shape_file:
- exp/asr_stats_raw_en_bpe5000_sp/train/speech_shape
- exp/asr_stats_raw_en_bpe5000_sp/train/text_shape.bpe
valid_shape_file:
- exp/asr_stats_raw_en_bpe5000_sp/valid/speech_shape
- exp/asr_stats_raw_en_bpe5000_sp/valid/text_shape.bpe
batch_type: numel
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_960_sp/wav.scp
- speech
- kaldi_ark
- - dump/raw/train_960_sp/text
- text
- text
valid_data_path_and_name_and_type:
- - dump/raw/dev/wav.scp
- speech
- kaldi_ark
- - 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: 0.0025
scheduler: warmuplr
scheduler_conf:
warmup_steps: 40000
token_list:
- <blank>
- <unk>
- ▁THE
- S
- ▁AND
- ▁OF
- ▁TO
- ▁A
- ▁IN
- ▁I
- ▁HE
- ▁THAT
- ▁WAS
- ED
- ▁IT
- ''''
- ▁HIS
- ING
- ▁YOU
- ▁WITH
- ▁FOR
- ▁HAD
- T
- ▁AS
- ▁HER
- ▁IS
- ▁BE
- ▁BUT
- ▁NOT
- ▁SHE
- D
- ▁AT
- ▁ON
- LY
- ▁HIM
- ▁THEY
- ▁ALL
- ▁HAVE
- ▁BY
- ▁SO
- ▁THIS
- ▁MY
- ▁WHICH
- ▁ME
- ▁SAID
- ▁FROM
- ▁ONE
- Y
- E
- ▁WERE
- ▁WE
- ▁NO
- N
- ▁THERE
- ▁OR
- ER
- ▁AN
- ▁WHEN
- ▁ARE
- ▁THEIR
- ▁WOULD
- ▁IF
- ▁WHAT
- ▁THEM
- ▁WHO
- ▁OUT
- M
- ▁DO
- ▁WILL
- ▁UP
- ▁BEEN
- P
- R
- ▁MAN
- ▁THEN
- ▁COULD
- ▁MORE
- C
- ▁INTO
- ▁NOW
- ▁VERY
- ▁YOUR
- ▁SOME
- ▁LITTLE
- ES
- ▁TIME
- RE
- ▁CAN
- ▁LIKE
- LL
- ▁ABOUT
- ▁HAS
- ▁THAN
- ▁DID
- ▁UPON
- ▁OVER
- IN
- ▁ANY
- ▁WELL
- ▁ONLY
- B
- ▁SEE
- ▁GOOD
- ▁OTHER
- ▁TWO
- L
- ▁KNOW
- ▁GO
- ▁DOWN
- ▁BEFORE
- A
- AL
- ▁OUR
- ▁OLD
- ▁SHOULD
- ▁MADE
- ▁AFTER
- ▁GREAT
- ▁DAY
- ▁MUST
- ▁COME
- ▁HOW
- ▁SUCH
- ▁CAME
- LE
- ▁WHERE
- ▁US
- ▁NEVER
- ▁THESE
- ▁MUCH
- ▁DE
- ▁MISTER
- ▁WAY
- G
- ▁S
- ▁MAY
- ATION
- ▁LONG
- OR
- ▁AM
- ▁FIRST
- ▁BACK
- ▁OWN
- ▁RE
- ▁AGAIN
- ▁SAY
- ▁MEN
- ▁WENT
- ▁HIMSELF
- ▁HERE
- NESS
- ▁THINK
- V
- IC
- ▁EVEN
- ▁THOUGHT
- ▁HAND
- ▁JUST
- ▁O
- ▁UN
- VE
- ION
- ▁ITS
- 'ON'
- ▁MAKE
- ▁MIGHT
- ▁TOO
- K
- ▁AWAY
- ▁LIFE
- TH
- ▁WITHOUT
- ST
- ▁THROUGH
- ▁MOST
- ▁TAKE
- ▁DON
- ▁EVERY
- F
- O
- ▁SHALL
- ▁THOSE
- ▁EYES
- AR
- ▁STILL
- ▁LAST
- ▁HOUSE
- ▁HEAD
- ABLE
- ▁NOTHING
- ▁NIGHT
- ITY
- ▁LET
- ▁MANY
- ▁OFF
- ▁BEING
- ▁FOUND
- ▁WHILE
- EN
- ▁SAW
- ▁GET
- ▁PEOPLE
- ▁FACE
- ▁YOUNG
- CH
- ▁UNDER
- ▁ONCE
- ▁TELL
- AN
- ▁THREE
- ▁PLACE
- ▁ROOM
- ▁YET
- ▁SAME
- IL
- US
- U
- ▁FATHER
- ▁RIGHT
- EL
- ▁THOUGH
- ▁ANOTHER
- LI
- RI
- ▁HEART
- IT
- ▁PUT
- ▁TOOK
- ▁GIVE
- ▁EVER
- ▁E
- ▁PART
- ▁WORK
- ERS
- ▁LOOK
- ▁NEW
- ▁KING
- ▁MISSUS
- ▁SIR
- ▁LOVE
- ▁MIND
- ▁LOOKED
- W
- RY
- ▁ASKED
- ▁LEFT
- ET
- ▁LIGHT
- CK
- ▁DOOR
- ▁MOMENT
- RO
- ▁WORLD
- ▁THINGS
- ▁HOME
- UL
- ▁THING
- LA
- ▁WHY
- ▁MOTHER
- ▁ALWAYS
- ▁FAR
- FUL
- ▁WATER
- CE
- IVE
- UR
- ▁HEARD
- ▁SOMETHING
- ▁SEEMED
- I
- LO
- ▁BECAUSE
- OL
- ▁END
- ▁TOLD
- ▁CON
- ▁YES
- ▁GOING
- ▁GOT
- RA
- IR
- ▁WOMAN
- ▁GOD
- EST
- TED
- ▁FIND
- ▁KNEW
- ▁SOON
- ▁EACH
- ▁SIDE
- H
- TON
- MENT
- ▁OH
- NE
- Z
- LING
- ▁AGAINST
- TER
- ▁NAME
- ▁MISS
- ▁QUITE
- ▁WANT
- ▁YEARS
- ▁FEW
- ▁BETTER
- ENT
- ▁HALF
- ▁DONE
- ▁ALSO
- ▁BEGAN
- ▁HAVING
- ▁ENOUGH
- IS
- ▁LADY
- ▁WHOLE
- LESS
- ▁BOTH
- ▁SEEN
- ▁SET
- ▁WHITE
- ▁COURSE
- IES
- ▁VOICE
- ▁CALLED
- ▁D
- ▁EX
- ATE
- ▁TURNED
- ▁GAVE
- ▁C
- ▁POOR
- MAN
- UT
- NA
- ▁DEAR
- ISH
- ▁GIRL
- ▁MORNING
- ▁BETWEEN
- LED
- ▁NOR
- IA
- ▁AMONG
- MA
- ▁
- ▁SMALL
- ▁REST
- ▁WHOM
- ▁FELT
- ▁HANDS
- ▁MYSELF
- ▁HIGH
- ▁M
- ▁HOWEVER
- ▁HERSELF
- ▁P
- CO
- ▁STOOD
- ID
- ▁KIND
- ▁HUNDRED
- AS
- ▁ROUND
- ▁ALMOST
- TY
- ▁SINCE
- ▁G
- AM
- ▁LA
- SE
- ▁BOY
- ▁MA
- ▁PERHAPS
- ▁WORDS
- ATED
- ▁HO
- X
- ▁MO
- ▁SAT
- ▁REPLIED
- ▁FOUR
- ▁ANYTHING
- ▁TILL
- ▁UNTIL
- ▁BLACK
- TION
- ▁CRIED
- RU
- TE
- ▁FACT
- ▁HELP
- ▁NEXT
- ▁LOOKING
- ▁DOES
- ▁FRIEND
- ▁LAY
- ANCE
- ▁POWER
- ▁BROUGHT
- VER
- ▁FIRE
- ▁KEEP
- PO
- FF
- ▁COUNTRY
- ▁SEA
- ▁WORD
- ▁CAR
- ▁DAYS
- ▁TOGETHER
- ▁IMP
- ▁REASON
- KE
- ▁INDEED
- TING
- ▁MATTER
- ▁FULL
- ▁TEN
- TIC
- ▁LAND
- ▁RATHER
- ▁AIR
- ▁HOPE
- ▁DA
- ▁OPEN
- ▁FEET
- ▁EN
- ▁FIVE
- ▁POINT
- ▁CO
- OM
- ▁LARGE
- ▁B
- ▁CL
- ME
- ▁GONE
- ▁CHILD
- INE
- GG
- ▁BEST
- ▁DIS
- UM
- ▁HARD
- ▁LORD
- OUS
- ▁WIFE
- ▁SURE
- ▁FORM
- DE
- ▁DEATH
- ANT
- ▁NATURE
- ▁BA
- ▁CARE
- ▁BELIEVE
- PP
- ▁NEAR
- ▁RO
- ▁RED
- ▁WAR
- IE
- ▁SPEAK
- ▁FEAR
- ▁CASE
- ▁TAKEN
- ▁ALONG
- ▁CANNOT
- ▁HEAR
- ▁THEMSELVES
- CI
- ▁PRESENT
- AD
- ▁MASTER
- ▁SON
- ▁THUS
- ▁LI
- ▁LESS
- ▁SUN
- ▁TRUE
- IM
- IOUS
- ▁THOUSAND
- ▁MONEY
- ▁W
- ▁BEHIND
- ▁CHILDREN
- ▁DOCTOR
- AC
- ▁TWENTY
- ▁WISH
- ▁SOUND
- ▁WHOSE
- ▁LEAVE
- ▁ANSWERED
- ▁THOU
- ▁DUR
- ▁HA
- ▁CERTAIN
- ▁PO
- ▁PASSED
- GE
- TO
- ▁ARM
- ▁LO
- ▁STATE
- ▁ALONE
- TA
- ▁SHOW
- ▁NEED
- ▁LIVE
- ND
- ▁DEAD
- ENCE
- ▁STRONG
- ▁PRE
- ▁TI
- ▁GROUND
- SH
- TI
- ▁SHORT
- IAN
- UN
- ▁PRO
- ▁HORSE
- MI
- ▁PRINCE
- ARD
- ▁FELL
- ▁ORDER
- ▁CALL
- AT
- ▁GIVEN
- ▁DARK
- ▁THEREFORE
- ▁CLOSE
- ▁BODY
- ▁OTHERS
- ▁SENT
- ▁SECOND
- ▁OFTEN
- ▁CA
- ▁MANNER
- MO
- NI
- ▁BRING
- ▁QUESTION
- ▁HOUR
- ▁BO
- AGE
- ▁ST
- ▁TURN
- ▁TABLE
- ▁GENERAL
- ▁EARTH
- ▁BED
- ▁REALLY
- ▁SIX
- 'NO'
- IST
- ▁BECOME
- ▁USE
- ▁READ
- ▁SE
- ▁VI
- ▁COMING
- ▁EVERYTHING
- ▁EM
- ▁ABOVE
- ▁EVENING
- ▁BEAUTIFUL
- ▁FEEL
- ▁RAN
- ▁LEAST
- ▁LAW
- ▁ALREADY
- ▁MEAN
- ▁ROSE
- WARD
- ▁ITSELF
- ▁SOUL
- ▁SUDDENLY
- ▁AROUND
- RED
- ▁ANSWER
- ICAL
- ▁RA
- ▁WIND
- ▁FINE
- ▁WON
- ▁WHETHER
- ▁KNOWN
- BER
- NG
- ▁TA
- ▁CAPTAIN
- ▁EYE
- ▁PERSON
- ▁WOMEN
- ▁SORT
- ▁ASK
- ▁BROTHER
- ▁USED
- ▁HELD
- ▁BIG
- ▁RETURNED
- ▁STRANGE
- ▁BU
- ▁PER
- ▁FREE
- ▁EITHER
- ▁WITHIN
- ▁DOUBT
- ▁YEAR
- ▁CLEAR
- ▁SIGHT
- ▁GRA
- ▁LOST
- ▁KEPT
- ▁F
- PE
- ▁BAR
- ▁TOWN
- ▁SLEEP
- ARY
- ▁HAIR
- ▁FRIENDS
- ▁DREAM
- ▁FELLOW
- PER
- ▁DEEP
- QUE
- ▁BECAME
- ▁REAL
- ▁PAST
- ▁MAKING
- RING
- ▁COMP
- ▁ACT
- ▁BAD
- HO
- STER
- ▁YE
- ▁MEANS
- ▁RUN
- MEN
- ▁DAUGHTER
- ▁SENSE
- ▁CITY
- ▁SOMETIMES
- ▁TOWARDS
- ▁ROAD
- ▁SP
- ▁LU
- ▁READY
- ▁FOOT
- ▁COLD
- ▁SA
- ▁LETTER
- ▁ELSE
- ▁MAR
- ▁STA
- BE
- ▁TRUTH
- ▁LE
- BO
- ▁BUSINESS
- CHE
- ▁JOHN
- ▁SUBJECT
- ▁COURT
- ▁IDEA
- ILY
- ▁RIVER
- ATING
- ▁FAMILY
- HE
- ▁DIDN
- ▁GLAD
- ▁SEVERAL
- IAL
- ▁UNDERSTAND
- ▁SC
- ▁POSSIBLE
- ▁DIFFERENT
- ▁RETURN
- ▁ARMS
- ▁LOW
- ▁HOLD
- ▁TALK
- ▁RU
- ▁WINDOW
- ▁INTEREST
- ▁SISTER
- SON
- ▁SH
- ▁BLOOD
- ▁SAYS
- ▁CAP
- ▁DI
- ▁HUMAN
- ▁CAUSE
- NCE
- ▁THANK
- ▁LATE
- GO
- ▁CUT
- ▁ACROSS
- ▁STORY
- NT
- ▁COUNT
- ▁ABLE
- DY
- LEY
- ▁NUMBER
- ▁STAND
- ▁CHURCH
- ▁THY
- ▁SUPPOSE
- LES
- BLE
- OP
- ▁EFFECT
- BY
- ▁K
- ▁NA
- ▁SPOKE
- ▁MET
- ▁GREEN
- ▁HUSBAND
- ▁RESPECT
- ▁PA
- ▁FOLLOWED
- ▁REMEMBER
- ▁LONGER
- ▁AGE
- ▁TAKING
- ▁LINE
- ▁SEEM
- ▁HAPPY
- LAND
- EM
- ▁STAY
- ▁PLAY
- ▁COMMON
- ▁GA
- ▁BOOK
- ▁TIMES
- ▁OBJECT
- ▁SEVEN
- QUI
- DO
- UND
- ▁FL
- ▁PRETTY
- ▁FAIR
- WAY
- ▁WOOD
- ▁REACHED
- ▁APPEARED
- ▁SWEET
- ▁FALL
- BA
- ▁PASS
- ▁SIGN
- ▁TREE
- IONS
- ▁GARDEN
- ▁ILL
- ▁ART
- ▁REMAIN
- ▁OPENED
- ▁BRIGHT
- ▁STREET
- ▁TROUBLE
- ▁PAIN
- ▁CONTINUED
- ▁SCHOOL
- OUR
- ▁CARRIED
- ▁SAYING
- HA
- ▁CHANGE
- ▁FOLLOW
- ▁GOLD
- ▁SW
- ▁FEELING
- ▁COMMAND
- ▁BEAR
- ▁CERTAINLY
- ▁BLUE
- ▁NE
- CA
- ▁WILD
- ▁ACCOUNT
- ▁OUGHT
- UD
- ▁T
- ▁BREATH
- ▁WANTED
- ▁RI
- ▁HEAVEN
- ▁PURPOSE
- ▁CHARACTER
- ▁RICH
- ▁PE
- ▁DRESS
- OS
- FA
- ▁TH
- ▁ENGLISH
- ▁CHANCE
- ▁SHIP
- ▁VIEW
- ▁TOWARD
- AK
- ▁JOY
- ▁JA
- ▁HAR
- ▁NEITHER
- ▁FORCE
- ▁UNCLE
- DER
- ▁PLAN
- ▁PRINCESS
- DI
- ▁CHIEF
- ▁HAT
- ▁LIVED
- ▁AB
- ▁VISIT
- ▁MOR
- TEN
- ▁WALL
- UC
- ▁MINE
- ▁PLEASURE
- ▁SMILE
- ▁FRONT
- ▁HU
- ▁DEAL
- OW
- ▁FURTHER
- GED
- ▁TRIED
- DA
- VA
- ▁NONE
- ▁ENTERED
- ▁QUEEN
- ▁PAY
- ▁EL
- ▁EXCEPT
- ▁SHA
- ▁FORWARD
- ▁EIGHT
- ▁ADDED
- ▁PUBLIC
- ▁EIGHTEEN
- ▁STAR
- ▁HAPPENED
- ▁LED
- ▁WALKED
- ▁ALTHOUGH
- ▁LATER
- ▁SPIRIT
- ▁WALK
- ▁BIT
- ▁MEET
- LIN
- ▁FI
- LT
- ▁MOUTH
- ▁WAIT
- ▁HOURS
- ▁LIVING
- ▁YOURSELF
- ▁FAST
- ▁CHA
- ▁HALL
- ▁BEYOND
- ▁BOAT
- ▁SECRET
- ENS
- ▁CHAIR
- RN
- ▁RECEIVED
- ▁CAT
- RESS
- ▁DESIRE
- ▁GENTLEMAN
- UGH
- ▁LAID
- EVER
- ▁OCCASION
- ▁WONDER
- ▁GU
- ▁PARTY
- DEN
- ▁FISH
- ▁SEND
- ▁NEARLY
- ▁TRY
- CON
- ▁SEEMS
- RS
- ▁BELL
- ▁BRA
- ▁SILENCE
- IG
- ▁GUARD
- ▁DIE
- ▁DOING
- ▁TU
- ▁COR
- ▁EARLY
- ▁BANK
- ▁FIGURE
- IF
- ▁ENGLAND
- ▁MARY
- ▁AFRAID
- LER
- ▁FO
- ▁WATCH
- ▁FA
- ▁VA
- ▁GRE
- ▁AUNT
- PED
- ▁SERVICE
- ▁JE
- ▁PEN
- ▁MINUTES
- ▁PAN
- ▁TREES
- NED
- ▁GLASS
- ▁TONE
- ▁PLEASE
- ▁FORTH
- ▁CROSS
- ▁EXCLAIMED
- ▁DREW
- ▁EAT
- ▁AH
- ▁GRAVE
- ▁CUR
- PA
- URE
- CENT
- ▁MILES
- ▁SOFT
- ▁AGO
- ▁POSITION
- ▁WARM
- ▁LENGTH
- ▁NECESSARY
- ▁THINKING
- ▁PICTURE
- ▁PI
- SHIP
- IBLE
- ▁HEAVY
- ▁ATTENTION
- ▁DOG
- ABLY
- ▁STANDING
- ▁NATURAL
- ▁APPEAR
- OV
- ▁CAUGHT
- VO
- ISM
- ▁SPRING
- ▁EXPERIENCE
- ▁PAT
- OT
- ▁STOPPED
- ▁REGARD
- ▁HARDLY
- ▁SELF
- ▁STRENGTH
- ▁GREW
- ▁KNIGHT
- ▁OPINION
- ▁WIDE
- ▁INSTEAD
- ▁SOUTH
- ▁TRANS
- ▁CORNER
- ▁LEARN
- ▁ISLAND
- ▁MI
- ▁THIRD
- ▁STE
- ▁STRAIGHT
- ▁TEA
- ▁BOUND
- ▁SEEING
- ▁JU
- ▁DINNER
- ▁BEAUTY
- ▁PEACE
- AH
- ▁REP
- ▁SILENT
- ▁CRE
- ALLY
- RIC
- ▁STEP
- ▁VER
- ▁JO
- GER
- ▁SITTING
- ▁THIRTY
- ▁SAVE
- ENED
- ▁GLANCE
- ▁REACH
- ▁ACTION
- ▁SAL
- ▁SAD
- ▁STONE
- ITIES
- ▁FRENCH
- ▁STRUCK
- ▁PAPER
- ▁WHATEVER
- ▁SUB
- ▁DISTANCE
- ▁WRONG
- ▁KNOWLEDGE
- ▁SAFE
- ▁SNOW
- ▁MUSIC
- ▁FIFTY
- RON
- ▁ATTEMPT
- ▁GOVERNMENT
- TU
- ▁CROWD
- ▁BESIDES
- ▁LOVED
- ▁BOX
- ▁DIRECTION
- ▁TRAIN
- ▁NORTH
- ▁THICK
- ▁GETTING
- AV
- ▁FLOOR
- ▁COMPANY
- ▁BLOW
- ▁PLAIN
- TRO
- ▁BESIDE
- ▁ROCK
- ▁IMMEDIATELY
- FI
- ▁SHADOW
- ▁SIT
- ORS
- ILE
- ▁DRINK
- ▁SPOT
- ▁DANGER
- ▁AL
- ▁SAINT
- ▁SLOWLY
- ▁PALACE
- IER
- ▁RESULT
- ▁PETER
- ▁FOREST
- ▁BELONG
- ▁SU
- ▁PAR
- RIS
- ▁TEARS
- ▁APPEARANCE
- ▁GATE
- BU
- ITION
- ▁QUICKLY
- ▁QUIET
- ▁LONDON
- ▁START
- ▁BROWN
- TRA
- KIN
- ▁CONSIDER
- ▁BATTLE
- ▁ANNE
- ▁PIECE
- ▁DIED
- ▁SUCCESS
- ▁LIPS
- ▁FILLED
- ▁FORGET
- ▁POST
- IFIED
- ▁MARGARET
- ▁FOOD
- HAM
- ▁PLEASANT
- ▁FE
- ▁EXPRESSION
- ▁POCKET
- ▁FRESH
- ▁WEAR
- TRI
- ▁BROKEN
- ▁LAUGHED
- GING
- ▁FOLLOWING
- WN
- IP
- ▁TOUCH
- ▁YOUTH
- ATIVE
- ▁LEG
- ▁WEEK
- ▁REMAINED
- ▁EASY
- NER
- RK
- ▁ENTER
- ▁FIGHT
- ▁PLACED
- ▁TRAVEL
- ▁SIMPLE
- ▁GIRLS
- ▁WAITING
- ▁STOP
- ▁WAVE
- AU
- ▁WISE
- ▁CAMP
- TURE
- UB
- ▁VE
- ▁OFFICE
- ▁GRAND
- ▁FIT
- ▁JUDGE
- UP
- MENTS
- ▁QUICK
- HI
- ▁FLO
- RIES
- VAL
- ▁COMFORT
- ▁PARTICULAR
- ▁STARTED
- ▁SUIT
- ▁NI
- ▁PALE
- ▁IMPOSSIBLE
- ▁HOT
- ▁CONVERSATION
- ▁SCENE
- ▁BOYS
- ▁WIN
- ▁BRE
- ▁SOCIETY
- ▁OUTSIDE
- ▁WRITE
- ▁EFFORT
- ▁TALKING
- ▁FORTUNE
- ▁NINE
- ▁WA
- ▁SINGLE
- ▁RULE
- ▁PORT
- ▁WINTER
- ▁CAST
- ▁CRA
- ▁HAPPEN
- ▁CRO
- ▁SHUT
- NING
- ▁GUN
- ▁NOBLE
- ▁BEGIN
- ▁PATH
- ▁SKY
- ▁WONDERFUL
- ▁SUDDEN
- ▁ARMY
- ▁CHE
- ▁WORTH
- ▁MOUNTAIN
- ▁MIN
- AG
- ▁FLU
- ▁GRACE
- ▁CHAPTER
- ▁BELOW
- ▁RING
- ▁TURNING
- ▁IRON
- ▁TOP
- ▁AFTERNOON
- ORY
- ▁EVIL
- ▁TRUST
- ▁BOW
- ▁TRI
- ▁SAIL
- ▁CONTENT
- ▁HORSES
- ITE
- ▁SILVER
- AP
- ▁LAD
- ▁RUNNING
- ▁HILL
- ▁BEGINNING
- ▁MAD
- ▁HABIT
- GRA
- ▁CLOTHES
- ▁MORROW
- ▁CRY
- ▁FASHION
- ▁PRESENCE
- ▁Z
- FE
- ▁ARRIVED
- ▁QUARTER
- ▁PERFECT
- ▁WO
- ▁TRA
- ▁USUAL
- ▁NECK
- ▁MARRIED
- ▁SEAT
- ▁WI
- ▁GAR
- ▁SAND
- ▁SHORE
- ▁GIVING
- NY
- ▁PROBABLY
- ▁MINUTE
- ▁EXPECT
- ▁DU
- ▁SHOT
- ▁INSTANT
- ▁DEGREE
- ▁COLOR
- ▁WEST
- RT
- ▁MARCH
- ▁BIRD
- ▁SHOWED
- ▁GREATER
- ▁SERIOUS
- ▁CARRY
- ▁COVERED
- ▁FORMER
- ▁LOUD
- ▁MOVED
- ▁MASS
- ▁SEEK
- ▁CHO
- GEN
- ▁ROMAN
- IB
- ▁MOON
- ▁BOARD
- ▁STREAM
- ▁EASILY
- ▁WISHED
- ▁SEARCH
- ▁COULDN
- ▁MONTHS
- ▁SICK
- LIE
- ▁DUTY
- ▁TWELVE
- ▁FAINT
- ▁STRANGER
- ▁SURPRISE
- ▁KILL
- ▁LEAVING
- ▁JOURNEY
- ▁SCARCELY
- ▁RAISED
- ▁SPEAKING
- ▁TERRIBLE
- ▁TOM
- ▁FIELD
- ▁GAME
- ▁QUA
- ▁PROMISE
- ▁LIE
- ▁CONDITION
- ▁TRO
- ▁PERSONAL
- ▁TALL
- ▁STICK
- ▁THREW
- ▁MARRY
- ▁VAN
- ▁BURN
- ▁ACCORDING
- ▁RISE
- ▁ATTACK
- ▁SWORD
- ▁GUESS
- ▁THOUGHTS
- ▁THIN
- ▁THROW
- ▁CALM
- SIDE
- ▁VILLAGE
- ▁DEN
- ▁ANXIOUS
- ▁MER
- GI
- ▁EXPECTED
- ▁BALL
- ▁ESPECIALLY
- ▁CHARGE
- ▁MEASURE
- ISE
- ▁NICE
- ▁TRYING
- ▁ALLOW
- ▁SHARP
- ▁BREAD
- ▁HONOUR
- ▁HONOR
- ▁ENTIRELY
- ▁BILL
- ▁BRI
- ▁WRITTEN
- ▁AR
- ▁BROKE
- ▁KILLED
- ▁MARK
- ▁VEN
- ▁LADIES
- ▁LEARNED
- ▁FLOWERS
- PLE
- ▁FORTY
- ▁OFFER
- ▁HAPPINESS
- ▁PRAY
- ▁CLASS
- ▁FER
- ▁PRINCIPLE
- GU
- ▁BOOKS
- ▁SHAPE
- ▁SUMMER
- ▁JACK
- ▁DRAW
- ▁GOLDEN
- ▁DECIDED
- ▁LEAD
- ▁UNLESS
- ▁HARM
- ▁LISTEN
- HER
- ▁SHOOK
- ▁INFLUENCE
- ▁PERFECTLY
- ▁MARRIAGE
- ▁BROAD
- ▁ESCAPE
- ▁STATES
- ▁MIDDLE
- ▁PLANT
- ▁MIL
- ▁MOVEMENT
- ▁NOISE
- ▁ENEMY
- ▁HISTORY
- ▁BREAK
- ROUS
- ▁UNDERSTOOD
- ▁LATTER
- FER
- ▁COMES
- ▁MERELY
- ▁SIMPLY
- WI
- ▁IMAGINE
- ▁LOWER
- ▁CONDUCT
- ▁BORN
- WA
- ▁YARD
- ▁KA
- ▁CLOSED
- ▁NOTE
- GA
- ▁STRA
- RAN
- ▁EXIST
- EV
- ▁SPEECH
- ▁BITTER
- JO
- ▁MAKES
- ▁GRASS
- ▁REPLY
- ▁CHANGED
- ▁MON
- ▁LYING
- ▁DANCE
- ▁FINALLY
- ▁AMERICAN
- ▁ENJOY
- ▁CONTAIN
- ▁MEANT
- USE
- ▁OBSERVED
- THER
- ▁LAUGH
- ▁AFTERWARDS
- ▁BEAT
- ▁RACE
- ▁EQUAL
- ▁RAIN
- PS
- ▁STEPS
- ▁BENEATH
- ▁TAIL
- ▁TASTE
- IO
- EY
- ▁CHAR
- ▁GE
- GN
- TIN
- ▁GROW
- ▁TE
- IANS
- ▁MOVE
- ▁REPEATED
- ▁DRIVE
- TUR
- ▁SI
- CLOCK
- ▁BRAVE
- ▁MADAME
- ▁LOT
- ▁CASTLE
- ▁HI
- AND
- ▁FUTURE
- ▁RELATION
- ▁SORRY
- ▁HEALTH
- ▁DICK
- ▁R
- ▁BUILDING
- ▁EDGE
- ▁BLESS
- ▁SPITE
- WE
- ▁MIS
- ▁PRISONER
- ▁ALLOWED
- ▁PH
- ▁CATCH
- MER
- ETH
- ▁COAT
- ▁COMPLETE
- ▁WOULDN
- ▁CREATURE
- ▁YELLOW
- ▁IMPORTANT
- ▁ADD
- ▁PASSING
- ▁DARKNESS
- ▁CARRIAGE
- ▁MILL
- ▁FIFTEEN
- NCY
- ▁HUNG
- ▁OB
- ▁PLEASED
- ▁SPREAD
- ▁CURIOUS
- ▁WORSE
- ▁CIRCUMSTANCES
- ▁GI
- LAR
- ▁CAL
- ▁HY
- ▁MERE
- ▁JANE
- ▁EAST
- BI
- ▁CUP
- ▁BLIND
- ▁PASSION
- ▁DISCOVERED
- ▁NOTICE
- ▁REPORT
- ▁SPACE
- ▁PRESENTLY
- ▁SORROW
- ▁PACK
- ▁DIN
- CY
- ▁DRY
- ▁ANCIENT
- ▁DRESSED
- ▁COVER
- ▁VO
- ▁EXISTENCE
- ▁EXACTLY
- ▁BEAST
- ▁PROPER
- ▁DROPPED
- ▁CLEAN
- ▁COLOUR
- ▁HOST
- ▁CHAMBER
- ▁FAITH
- LET
- ▁DETERMINED
- ▁PRIEST
- ▁STORM
- ▁SKIN
- ▁DARE
- ▁PERSONS
- ▁PICK
- ▁NARROW
- ▁SUPPORT
- ▁PRIVATE
- ▁SMILED
- ▁COUSIN
- ▁DRAWING
- ▁ATTEND
- ▁COOK
- ▁PREVENT
- ▁VARIOUS
- ▁BLA
- ▁FIXED
- ▁WEAK
- THE
- ▁HOLE
- ▁BOTTOM
- ▁NOBODY
- ADE
- ▁LEGS
- ITCH
- ▁INDIVIDUAL
- ▁EARS
- LIKE
- ▁ADVANTAGE
- ▁FRANCE
- ▁BON
- ▁WINE
- ▁LIVES
- OD
- ▁WALLS
- ▁TIRED
- ▁SHOP
- ▁ANIMAL
- ▁CRU
- ▁WROTE
- ▁ROYAL
- ▁CONSIDERED
- ▁MORAL
- ▁COMPANION
- ▁LOSE
- ▁ISN
- ▁BAG
- ▁LAKE
- ▁INTER
- ▁COM
- ▁LETTERS
- ▁LUCK
- ▁EAR
- ▁GERMAN
- ▁PET
- ▁SAKE
- ▁DROP
- ▁PAID
- ▁BREAKFAST
- ▁LABOR
- ▁DESERT
- ▁DECLARED
- ▁HUM
- ▁STUDY
- ▁INSTANCE
- ONE
- ▁SOMEWHAT
- ▁CLOTH
- ▁SPECIAL
- ▁COLONEL
- ▁SONG
- ▁MAIN
- ▁VALUE
- ▁PROUD
- ▁EXPRESS
- ▁NATION
- ▁HANDSOME
- ▁CONFESS
- ▁PU
- ▁PASSAGE
- ▁PERIOD
- ▁CUSTOM
- ▁HURT
- ▁SHOULDER
- ▁CHRIST
- ZA
- ▁RECEIVE
- ▁DIFFICULT
- ▁DEPEND
- ▁MEETING
- ▁CHI
- ▁GEN
- LIGHT
- ▁BELIEVED
- ▁SOCIAL
- ▁DIFFICULTY
- ▁GREATEST
- ▁DRAWN
- ▁GRANT
- ▁BIRDS
- ▁ANGRY
- ▁HEAT
- UFF
- ▁DUE
- ▁PLACES
- ▁SIN
- ▁COURAGE
- ▁EVIDENTLY
- ▁GENTLE
- ▁CRUEL
- ▁GEORGE
- ▁GRI
- ▁SERVANT
- ▁U
- ▁PURE
- OOK
- ▁KNOWS
- ▁KNOWING
- LF
- ▁WRITING
- ▁REMEMBERED
- ▁CU
- ▁HOLDING
- ▁TENDER
- ▁QUI
- ▁BURST
- ▁SURELY
- IGN
- ▁VALLEY
- ▁FU
- ▁BUTTER
- ▁SPOKEN
- ▁STORE
- ▁DISC
- ▁CHRISTIAN
- ▁PARIS
- ▁HENRY
- ▁FINISHED
- ▁PROVE
- ▁FOOL
- ▁SOLDIERS
- ▁LANGUAGE
- ▁INSIDE
- ▁BAN
- ▁FALLEN
- ROW
- ▁MAL
- ▁BABY
- ▁SITUATION
- ▁WATCHED
- ANS
- ▁RUIN
- ▁GENTLEMEN
- ▁FRO
- ▁FANCY
- ▁ACCEPT
- ▁SEASON
- ▁OURSELVES
- ▁SAN
- ▁SPEED
- IZED
- ▁COOL
- ▁SERVE
- ▁VESSEL
- ▁WILLIAM
- ▁OBLIGED
- ▁GROUP
- FORM
- ▁GOES
- UOUS
- ▁LEAVES
- ▁PECULIAR
- ▁NEWS
- ▁VAIN
- ▁EVERYBODY
- ▁PIN
- UG
- ▁FORGOTTEN
- ▁FRA
- GAN
- ▁CAREFULLY
- ▁FLASH
- UCH
- ▁FUR
- ▁MURDER
- ▁DELIGHT
- ▁WAITED
- ▁RENDER
- ▁PROPERTY
- ▁NOTICED
- ▁ROLL
- ▁KNOCK
- ▁EARNEST
- KI
- ▁HONEST
- ▁PROMISED
- ▁BAL
- AW
- ▁WALKING
- ANG
- ▁SQUARE
- ▁QUIETLY
- ▁CLOUD
- WOOD
- ▁FORMED
- ▁HIGHER
- ▁BUILT
- ▁FATE
- ▁TEACH
- MY
- ▁FALSE
- ▁YORK
- ▁DUST
- ▁CLIMB
- ▁FOND
- ▁GROWN
- ▁DESCEND
- ▁RAG
- ▁FRUIT
- ▁GENERALLY
- ▁OFFERED
- ▁ER
- ▁NURSE
- POSE
- ▁SPENT
- ▁JOIN
- ▁STATION
- ▁MEANING
- ▁SMOKE
- HOOD
- ▁ROUGH
- JU
- ▁LIKELY
- ▁SURFACE
- ▁KE
- ▁
📄 許可證
本項目採用CC BY 4.0許可證。
Voice Activity Detection
MIT
基於pyannote.audio 2.1版本的語音活動檢測模型,用於識別音頻中的語音活動時間段
語音識別
V
pyannote
7.7M
181
Wav2vec2 Large Xlsr 53 Portuguese
Apache-2.0
這是一個針對葡萄牙語語音識別任務微調的XLSR-53大模型,基於Common Voice 6.1數據集訓練,支持葡萄牙語語音轉文本。
語音識別 其他
W
jonatasgrosman
4.9M
32
Whisper Large V3
Apache-2.0
Whisper是由OpenAI提出的先進自動語音識別(ASR)和語音翻譯模型,在超過500萬小時的標註數據上訓練,具有強大的跨數據集和跨領域泛化能力。
語音識別 支持多種語言
W
openai
4.6M
4,321
Whisper Large V3 Turbo
MIT
Whisper是由OpenAI開發的最先進的自動語音識別(ASR)和語音翻譯模型,經過超過500萬小時標記數據的訓練,在零樣本設置下展現出強大的泛化能力。
語音識別
Transformers 支持多種語言

W
openai
4.0M
2,317
Wav2vec2 Large Xlsr 53 Russian
Apache-2.0
基於facebook/wav2vec2-large-xlsr-53模型微調的俄語語音識別模型,支持16kHz採樣率的語音輸入
語音識別 其他
W
jonatasgrosman
3.9M
54
Wav2vec2 Large Xlsr 53 Chinese Zh Cn
Apache-2.0
基於facebook/wav2vec2-large-xlsr-53模型微調的中文語音識別模型,支持16kHz採樣率的語音輸入。
語音識別 中文
W
jonatasgrosman
3.8M
110
Wav2vec2 Large Xlsr 53 Dutch
Apache-2.0
基於facebook/wav2vec2-large-xlsr-53微調的荷蘭語語音識別模型,在Common Voice和CSS10數據集上訓練,支持16kHz音頻輸入。
語音識別 其他
W
jonatasgrosman
3.0M
12
Wav2vec2 Large Xlsr 53 Japanese
Apache-2.0
基於facebook/wav2vec2-large-xlsr-53模型微調的日語語音識別模型,支持16kHz採樣率的語音輸入
語音識別 日語
W
jonatasgrosman
2.9M
33
Mms 300m 1130 Forced Aligner
基於Hugging Face預訓練模型的文本與音頻強制對齊工具,支持多種語言,內存效率高
語音識別
Transformers 支持多種語言

M
MahmoudAshraf
2.5M
50
Wav2vec2 Large Xlsr 53 Arabic
Apache-2.0
基於facebook/wav2vec2-large-xlsr-53微調的阿拉伯語語音識別模型,在Common Voice和阿拉伯語語音語料庫上訓練
語音識別 阿拉伯語
W
jonatasgrosman
2.3M
37
精選推薦AI模型
Llama 3 Typhoon V1.5x 8b Instruct
專為泰語設計的80億參數指令模型,性能媲美GPT-3.5-turbo,優化了應用場景、檢索增強生成、受限生成和推理任務
大型語言模型
Transformers 支持多種語言

L
scb10x
3,269
16
Cadet Tiny
Openrail
Cadet-Tiny是一個基於SODA數據集訓練的超小型對話模型,專為邊緣設備推理設計,體積僅為Cosmo-3B模型的2%左右。
對話系統
Transformers 英語

C
ToddGoldfarb
2,691
6
Roberta Base Chinese Extractive Qa
基於RoBERTa架構的中文抽取式問答模型,適用於從給定文本中提取答案的任務。
問答系統 中文
R
uer
2,694
98