metadata
license: cc-by-nc-sa-4.0
datasets:
- espnet/yodas
- facebook/voxpopuli
- facebook/multilingual_librispeech
- google/fleurs
- openslr/librispeech_asr
- speechcolab/gigaspeech
- cheulyop/ksponspeech
- espnet/mms_ulab_v2
library_name: espnet
tags:
- espnet
- audio
- speech
- multilingual
- automatic-speech-recognition
language:
- multilingual
- aaa
- aab
- aac
- aad
- aaf
- aai
- aal
- aao
- aap
- aar
- aau
- aaw
- aaz
- aba
- abh
- abi
- abm
- abn
- abo
- abr
- abs
- abt
- abu
- abz
- aca
- acd
- ace
- acf
- ach
- acm
- acn
- acq
- acr
- acu
- acv
- acw
- acz
- ada
- add
- ade
- adh
- adi
- adj
- adl
- adn
- ado
- adq
- adx
- ady
- adz
- aeb
- aec
- aee
- ael
- aeu
- aey
- aez
- afb
- afe
- afi
- afo
- afr
- afu
- afz
- agb
- agc
- agd
- age
- agf
- agg
- agh
- agi
- agl
- agn
- agq
- agr
- ags
- agt
- agu
- agw
- agy
- aha
- ahb
- ahg
- ahk
- ahl
- ahp
- ahr
- ahs
- aia
- aif
- aii
- aik
- aim
- aio
- aiw
- aix
- ajg
- aji
- akb
- akc
- akd
- ake
- akf
- akg
- akh
- aki
- akl
- akp
- akq
- akr
- aks
- akt
- akw
- ala
- ald
- ale
- alf
- alh
- alj
- alk
- all
- aln
- alp
- alq
- als
- alt
- alu
- alw
- alx
- aly
- alz
- amb
- amc
- ame
- amf
- amh
- ami
- amk
- amm
- amn
- amo
- amr
- amt
- amu
- anc
- anf
- anj
- ank
- anl
- anm
- ann
- ano
- anp
- anr
- anu
- anv
- anw
- anx
- any
- aoe
- aof
- aog
- aoi
- aoj
- aol
- aom
- aon
- aot
- aoz
- apb
- apc
- apd
- ape
- apj
- apm
- apn
- app
- apr
- apt
- apu
- apw
- apy
- apz
- aqg
- aqm
- aqt
- arb
- are
- arg
- arh
- arl
- arn
- aro
- arp
- arq
- arr
- arv
- arw
- arx
- ary
- arz
- asa
- asb
- asc
- asi
- ask
- asm
- aso
- asr
- ass
- asu
- asy
- ata
- atb
- atd
- atg
- ati
- atk
- ato
- atp
- atq
- ats
- att
- atu
- aty
- auc
- aug
- aui
- auk
- aul
- aun
- aup
- auq
- auu
- auy
- ava
- avd
- avi
- avl
- avn
- avt
- avu
- awa
- awb
- awe
- awi
- awn
- awu
- aww
- axk
- ayb
- ayg
- ayi
- ayn
- ayo
- ayp
- ayr
- ayt
- ayu
- ayz
- azb
- azd
- azg
- azj
- azm
- azt
- azz
- baa
- bab
- bac
- bag
- bam
- ban
- bao
- bap
- bar
- bas
- bau
- bav
- baw
- bax
- bba
- bbb
- bbc
- bbf
- bbi
- bbk
- bbo
- bbp
- bbq
- bbr
- bbt
- bbu
- bbv
- bbw
- bby
- bca
- bcc
- bcf
- bcg
- bci
- bcj
- bcl
- bcn
- bco
- bcp
- bcq
- bcr
- bcs
- bcv
- bcw
- bcy
- bcz
- bda
- bdb
- bdd
- bde
- bdh
- bdi
- bdl
- bdm
- bdq
- bdu
- bdv
- bdw
- bea
- bec
- bee
- bef
- beh
- bei
- bej
- bek
- bel
- bem
- ben
- beo
- bep
- beq
- bet
- beu
- bev
- bew
- bex
- bey
- bez
- bfa
- bfb
- bfd
- bfe
- bfg
- bfh
- bfj
- bfm
- bfo
- bfq
- bfr
- bfs
- bft
- bfu
- bfw
- bfy
- bfz
- bga
- bgc
- bgd
- bge
- bgf
- bgg
- bgi
- bgj
- bgn
- bgp
- bgq
- bgr
- bgs
- bgt
- bgv
- bgw
- bgx
- bgz
- bha
- bhb
- bhd
- bhf
- bhg
- bhh
- bhi
- bhj
- bhl
- bho
- bhp
- bhq
- bhr
- bhs
- bht
- bhu
- bhw
- bhx
- bhy
- bhz
- bib
- bid
- bif
- big
- bil
- bim
- bin
- bio
- bip
- bis
- bit
- biu
- biv
- bix
- biy
- biz
- bja
- bjc
- bje
- bjg
- bjh
- bji
- bjj
- bjk
- bjn
- bjo
- bjp
- bjr
- bjt
- bjx
- bjz
- bka
- bkc
- bkd
- bkg
- bkk
- bkl
- bkm
- bkq
- bkr
- bks
- bku
- bkv
- bkw
- bkx
- bky
- bla
- blb
- blc
- ble
- blf
- blh
- bli
- blk
- blm
- blo
- blq
- blr
- blt
- blw
- bly
- blz
- bma
- bmb
- bmd
- bmf
- bmi
- bmj
- bmk
- bmm
- bmq
- bmr
- bmu
- bmv
- bni
- bnj
- bnm
- bnn
- bno
- bnp
- bns
- bnv
- bnx
- boa
- bob
- bod
- bof
- boh
- bol
- bom
- bon
- boo
- boq
- bor
- bos
- bot
- bou
- bov
- box
- boz
- bpa
- bpe
- bpn
- bpp
- bpr
- bps
- bpu
- bpv
- bpw
- bpx
- bpy
- bpz
- bqa
- bqc
- bqg
- bqh
- bqi
- bqj
- bqo
- bqr
- bqs
- bqt
- bqv
- bqw
- bqx
- bra
- brb
- brd
- bre
- brf
- brg
- brh
- bri
- brl
- brp
- brq
- brr
- brt
- bru
- brv
- brx
- bsc
- bse
- bsf
- bsh
- bsi
- bsk
- bsn
- bsp
- bsq
- bss
- bst
- bsy
- bta
- btd
- bte
- btg
- btm
- bts
- btt
- btu
- btx
- bub
- bud
- buf
- bug
- buh
- bui
- buj
- buk
- bul
- bum
- bun
- buo
- bus
- buu
- buw
- bux
- buz
- bva
- bvc
- bvd
- bvh
- bvi
- bvm
- bvr
- bvu
- bvw
- bvz
- bwd
- bwe
- bwf
- bwi
- bwm
- bwo
- bwq
- bwr
- bws
- bwt
- bwu
- bww
- bwx
- bxa
- bxb
- bxg
- bxh
- bxk
- bxl
- bxq
- bxr
- bxs
- bya
- byc
- byd
- bye
- byj
- byn
- byo
- byp
- bys
- byv
- byx
- byz
- bza
- bzd
- bze
- bzf
- bzh
- bzi
- bzu
- bzv
- bzw
- bzx
- bzy
- bzz
- caa
- cab
- cac
- cae
- caf
- cag
- cak
- can
- cao
- cap
- caq
- car
- cas
- cat
- cav
- cax
- cay
- caz
- cbc
- cbd
- cbg
- cbi
- cbj
- cbk
- cbn
- cbo
- cbr
- cbs
- cbt
- cbu
- cbv
- cce
- ccg
- cch
- ccj
- ccl
- cco
- ccp
- cde
- cdf
- cdh
- cdi
- cdj
- cdm
- cdn
- cdo
- cdr
- cdz
- ceb
- ceg
- cek
- ces
- cfa
- cfd
- cfg
- cfm
- cgg
- cgk
- chb
- chd
- che
- chf
- chj
- chk
- chl
- cho
- chp
- chq
- chr
- chw
- chx
- chy
- cia
- cib
- cih
- cik
- cin
- ciw
- cja
- cje
- cjk
- cjm
- cjo
- cjv
- ckb
- ckh
- ckl
- cko
- ckt
- cku
- ckx
- cky
- cla
- clc
- cld
- cle
- cli
- clj
- clk
- cll
- clo
- clt
- clu
- cly
- cma
- cme
- cmn
- cmo
- cmr
- cna
- cnb
- cnc
- cnh
- cni
- cnk
- cnl
- cnq
- cns
- cnt
- cnw
- cob
- coc
- cod
- cof
- cog
- coh
- coj
- com
- con
- cos
- cou
- cov
- cox
- coz
- cpa
- cpx
- cqd
- cra
- crc
- crh
- crj
- crk
- crn
- cro
- crq
- crt
- crv
- crw
- crx
- cry
- csa
- csh
- csk
- cso
- csy
- cta
- ctd
- cte
- ctg
- ctl
- cto
- ctp
- ctt
- ctu
- ctz
- cua
- cub
- cuc
- cui
- cuk
- cul
- cut
- cuv
- cux
- cvg
- cvn
- cya
- cyb
- cym
- cyo
- czh
- czn
- czt
- daa
- dad
- dag
- dai
- dak
- dan
- dao
- daq
- das
- dav
- daw
- dax
- dbb
- dbd
- dbi
- dbj
- dbm
- dbn
- dbq
- dbv
- dby
- dcc
- dde
- ddg
- ddn
- dee
- def
- deg
- deh
- dei
- dem
- der
- deu
- dez
- dga
- dgc
- dgd
- dge
- dgg
- dgh
- dgi
- dgo
- dgr
- dgx
- dgz
- dhd
- dhg
- dhi
- dhm
- dhn
- dho
- dhv
- dhw
- dia
- dib
- did
- dig
- dih
- dij
- dik
- dil
- dim
- dio
- dip
- dir
- dis
- diu
- div
- diw
- diz
- djc
- dje
- djk
- djm
- djn
- djo
- djr
- dka
- dks
- dkx
- dln
- dma
- dme
- dmg
- dmo
- dmr
- dms
- dmw
- dna
- dnd
- dni
- dnj
- dnn
- dnw
- dny
- doa
- dob
- dof
- doo
- dop
- dor
- dos
- dot
- dow
- dox
- doy
- doz
- drd
- dre
- drg
- dri
- drs
- dru
- dry
- dsh
- dsn
- dsq
- dta
- dtb
- dtm
- dtp
- dts
- dty
- dua
- dub
- duc
- due
- dug
- duh
- dun
- duq
- dur
- dus
- duu
- duv
- duw
- dva
- dwa
- dwr
- dwu
- dww
- dwy
- dwz
- dya
- dyg
- dyi
- dyo
- dyu
- dza
- dzg
- dzl
- dzo
- ebo
- ebr
- ebu
- efi
- ega
- ego
- eip
- eit
- eja
- eka
- ekg
- ekl
- ekp
- ekr
- eky
- elk
- ell
- elm
- ema
- emb
- eme
- emg
- emk
- emn
- emp
- ems
- ena
- enb
- end
- eng
- enl
- enn
- enq
- env
- enx
- eot
- epi
- erg
- erh
- erk
- ert
- ese
- esg
- esh
- esi
- esk
- ess
- esu
- etn
- eto
- etr
- ets
- etu
- etx
- eus
- eve
- evn
- ewe
- ewo
- eyo
- eza
- eze
- faa
- fai
- fak
- fal
- fan
- fap
- far
- fat
- fay
- ffm
- fie
- fij
- fin
- fir
- fla
- fli
- fll
- flr
- fod
- foi
- fon
- for
- fqs
- fra
- frc
- frd
- fry
- fub
- fuc
- fue
- fuf
- fuh
- fun
- fuq
- fut
- fuu
- fuv
- fuy
- fvr
- fwe
- gaa
- gab
- gad
- gae
- gaf
- gah
- gai
- gaj
- gaq
- gar
- gas
- gau
- gaw
- gax
- gaz
- gbe
- gbg
- gbh
- gbi
- gbk
- gbl
- gbm
- gbn
- gbo
- gbr
- gbv
- gby
- gbz
- gcd
- gcf
- gcn
- gcr
- gdb
- gde
- gdf
- gdl
- gdn
- gdr
- gdu
- gdx
- gea
- geb
- gec
- ged
- geg
- gej
- gek
- gel
- gew
- gfk
- gga
- ggb
- ggg
- ggu
- ggw
- ghe
- ghk
- ghl
- ghn
- ghr
- ghs
- gia
- gid
- gig
- gil
- gim
- gis
- git
- giw
- giz
- gjk
- gjn
- gju
- gkn
- gkp
- gla
- gle
- glg
- glh
- glj
- glk
- glo
- glr
- glw
- gmb
- gmm
- gmv
- gmz
- gna
- gnb
- gnd
- gng
- gni
- gnk
- gnm
- gnn
- gno
- gnu
- gnw
- goa
- gof
- gog
- goj
- gok
- gol
- gom
- gop
- gor
- gou
- gow
- gox
- goz
- gpa
- gqa
- gra
- grd
- grh
- gri
- grj
- gro
- grs
- grt
- gru
- grv
- grx
- gry
- gsw
- gua
- gub
- guc
- gud
- gue
- guf
- gug
- guh
- gui
- guj
- guk
- gul
- gum
- gun
- guo
- gup
- guq
- gur
- gut
- guu
- guw
- gux
- guz
- gvc
- gvf
- gvj
- gvn
- gvo
- gvp
- gvr
- gvs
- gwa
- gwd
- gwi
- gwn
- gwr
- gwt
- gww
- gxx
- gya
- gyd
- gym
- gyr
- gyz
- haa
- hac
- had
- hae
- hag
- hah
- haj
- hak
- hal
- haq
- har
- has
- hat
- hau
- hav
- haw
- hay
- haz
- hbb
- hbn
- hca
- hch
- hdn
- hdy
- hea
- heb
- hed
- heg
- heh
- hei
- her
- hgm
- hgw
- hia
- hid
- hif
- hig
- hii
- hil
- hin
- hio
- hix
- hkk
- hla
- hlb
- hld
- hlt
- hmb
- hmd
- hmg
- hmj
- hml
- hmo
- hmr
- hms
- hmt
- hmw
- hmz
- hna
- hnd
- hne
- hni
- hnj
- hnn
- hno
- hns
- hoa
- hoc
- hoe
- hoj
- hol
- hoo
- hop
- hot
- how
- hoy
- hra
- hre
- hrm
- hru
- hrv
- hsn
- hto
- hts
- hub
- huc
- hue
- huf
- huh
- hui
- hul
- hum
- hun
- hup
- hur
- hus
- hut
- huv
- hux
- hve
- hvn
- hvv
- hwo
- hye
- hyw
- iai
- ian
- iar
- iba
- ibb
- ibd
- ibg
- ibl
- ibm
- ibo
- iby
- ica
- ich
- icr
- ida
- idi
- idu
- ifa
- ifb
- ife
- ifk
- ifm
- ifu
- ify
- igb
- ige
- igl
- ign
- ihp
- iii
- ijc
- ijj
- ijn
- ijs
- ike
- iki
- ikk
- iko
- ikt
- ikw
- ikx
- ilb
- ilk
- ilo
- ilp
- ilu
- imo
- ind
- inj
- ino
- int
- ior
- iow
- ipo
- iqu
- iqw
- iri
- irk
- irn
- irr
- iru
- irx
- iry
- isd
- ish
- isi
- isk
- isl
- isn
- iso
- isu
- ita
- itd
- ite
- iti
- ito
- itr
- its
- itt
- itv
- ity
- itz
- ium
- ivb
- ivv
- iwm
- iws
- ixl
- iyo
- iyx
- izr
- izz
- jaa
- jab
- jac
- jad
- jaf
- jam
- jao
- jaq
- jat
- jav
- jax
- jbj
- jbm
- jbu
- jda
- jdg
- jeb
- jeh
- jei
- jen
- jer
- jge
- jgk
- jib
- jic
- jid
- jig
- jio
- jit
- jiu
- jiv
- jiy
- jkp
- jkr
- jku
- jle
- jma
- jmb
- jmc
- jmd
- jmi
- jml
- jmn
- jmr
- jms
- jmx
- jna
- jnd
- jni
- jnj
- jnl
- jns
- job
- jog
- jow
- jpn
- jqr
- jra
- jrt
- jru
- jub
- juk
- jul
- jum
- jun
- juo
- jup
- jwi
- jya
- kaa
- kab
- kac
- kad
- kai
- kaj
- kak
- kal
- kam
- kan
- kao
- kap
- kaq
- kas
- kat
- kay
- kaz
- kbb
- kbc
- kbd
- kbh
- kbj
- kbl
- kbm
- kbo
- kbp
- kbq
- kbr
- kbv
- kbx
- kby
- kbz
- kcc
- kcd
- kce
- kcf
- kcg
- kch
- kci
- kcj
- kck
- kcl
- kcq
- kcr
- kcs
- kcv
- kcx
- kdd
- kde
- kdh
- kdi
- kdj
- kdl
- kdm
- kdp
- kdq
- kdt
- kdu
- kdx
- kdy
- kdz
- kea
- keb
- kee
- kef
- kei
- kej
- kek
- kel
- kem
- ken
- keo
- kep
- ker
- keu
- kev
- kex
- key
- kez
- kfa
- kfb
- kfc
- kfd
- kfe
- kff
- kfg
- kfh
- kfi
- kfk
- kfm
- kfo
- kfp
- kfq
- kfr
- kfs
- kft
- kfu
- kfv
- kfx
- kfy
- kfz
- kga
- kgb
- kge
- kgj
- kgk
- kgo
- kgp
- kgq
- kgr
- kgy
- kha
- khb
- khc
- khe
- khg
- khj
- khk
- khl
- khm
- khn
- khq
- khr
- khs
- kht
- khu
- khw
- khy
- khz
- kia
- kib
- kic
- kid
- kie
- kif
- kih
- kij
- kik
- kil
- kin
- kio
- kip
- kir
- kis
- kit
- kiu
- kiw
- kix
- kjb
- kjc
- kjd
- kje
- kjg
- kji
- kjl
- kjo
- kjp
- kjq
- kjr
- kjs
- kjt
- kkc
- kkd
- kkf
- kkh
- kkj
- kkk
- kkn
- kks
- kku
- kky
- kkz
- kla
- klb
- kle
- klg
- kli
- klk
- klo
- klq
- klr
- kls
- klu
- klv
- klw
- klx
- klz
- kma
- kmb
- kmc
- kmh
- kmi
- kmj
- kmk
- kml
- kmm
- kmn
- kmo
- kmp
- kmq
- kmr
- kms
- kmt
- kmu
- kmw
- kmy
- kmz
- kna
- knc
- knd
- kne
- knf
- kng
- kni
- knj
- knk
- knl
- knm
- knn
- kno
- knp
- knt
- knu
- knv
- knw
- knx
- kny
- knz
- kod
- koe
- kof
- koh
- koi
- kol
- koo
- kor
- kos
- kot
- kow
- kpa
- kpb
- kpc
- kph
- kpj
- kpk
- kpl
- kpm
- kpo
- kpq
- kpr
- kps
- kpw
- kpx
- kpz
- kqa
- kqb
- kqc
- kqe
- kqf
- kqi
- kqj
- kqk
- kql
- kqm
- kqn
- kqo
- kqp
- kqs
- kqw
- kqy
- kra
- krc
- krf
- krh
- kri
- krj
- krn
- krp
- krr
- krs
- kru
- krv
- krw
- krx
- ksb
- ksd
- ksf
- ksg
- ksi
- ksj
- ksm
- ksn
- ksp
- kss
- kst
- ksu
- ksv
- ksw
- ktb
- ktc
- ktf
- ktm
- ktn
- ktp
- ktu
- ktv
- kty
- ktz
- kua
- kub
- kud
- kue
- kuh
- kui
- kuj
- kul
- kun
- kup
- kus
- kuy
- kvb
- kvd
- kvf
- kvg
- kvi
- kvj
- kvl
- kvm
- kvn
- kvo
- kvq
- kvr
- kvt
- kvu
- kvv
- kvw
- kvx
- kvy
- kwa
- kwb
- kwc
- kwd
- kwe
- kwf
- kwg
- kwi
- kwj
- kwk
- kwl
- kwn
- kwo
- kws
- kwt
- kwu
- kwv
- kwx
- kxb
- kxc
- kxf
- kxh
- kxj
- kxm
- kxn
- kxp
- kxv
- kxw
- kxx
- kxz
- kyb
- kyc
- kye
- kyf
- kyg
- kyh
- kyk
- kyo
- kyq
- kys
- kyu
- kyv
- kyy
- kyz
- kza
- kzc
- kzf
- kzi
- kzm
- kzq
- kzr
- kzs
- laa
- lac
- lad
- lae
- lag
- lai
- laj
- lal
- lam
- lan
- lao
- lar
- las
- law
- lax
- lbf
- lbj
- lbk
- lbm
- lbn
- lbo
- lbq
- lbr
- lbu
- lbw
- lbx
- lcc
- lch
- lcm
- lcp
- ldb
- ldg
- ldi
- ldj
- ldk
- ldl
- ldm
- ldo
- ldp
- ldq
- lea
- lec
- led
- lee
- lef
- leh
- lek
- lel
- lem
- lep
- leq
- ler
- les
- leu
- lev
- lew
- lex
- lez
- lga
- lgg
- lgl
- lgm
- lgq
- lgr
- lgt
- lgu
- lhi
- lhl
- lhm
- lhp
- lht
- lhu
- lia
- lic
- lie
- lif
- lig
- lih
- lik
- lil
- lin
- lip
- liq
- lir
- lis
- lit
- liu
- liw
- liz
- lje
- ljp
- lkh
- lki
- lkn
- lkr
- lkt
- lky
- lla
- llc
- lle
- llg
- lln
- llp
- llu
- lma
- lmd
- lme
- lmg
- lmi
- lmk
- lml
- lmn
- lmp
- lmu
- lmx
- lmy
- lna
- lnd
- lns
- lnu
- loa
- lob
- loe
- log
- loh
- lok
- lol
- lom
- lop
- loq
- lor
- los
- lot
- loy
- loz
- lpa
- lpn
- lpo
- lra
- lrc
- lri
- lrk
- lrl
- lrm
- lro
- lse
- lsh
- lsi
- lsm
- lsr
- lti
- ltz
- lua
- lub
- luc
- lue
- lug
- lui
- luj
- lul
- lum
- lun
- luo
- lup
- lur
- lus
- luz
- lva
- lvk
- lvs
- lwg
- lwl
- lwo
- lyg
- lyn
- lzz
- maa
- mab
- mad
- mae
- maf
- mag
- mah
- mai
- maj
- mak
- mal
- mam
- mar
- mas
- mat
- mau
- mav
- maw
- max
- maz
- mbb
- mbc
- mbd
- mbf
- mbh
- mbi
- mbj
- mbl
- mbm
- mbo
- mbp
- mbq
- mbs
- mbt
- mbu
- mbv
- mbx
- mbz
- mca
- mcc
- mcd
- mcf
- mch
- mck
- mcn
- mco
- mcp
- mcq
- mcr
- mcs
- mct
- mcu
- mcw
- mda
- mdb
- mdd
- mde
- mdh
- mdj
- mdk
- mdm
- mdn
- mdr
- mds
- mdt
- mdu
- mdw
- mdy
- mea
- med
- mef
- meh
- mej
- mek
- men
- mep
- mer
- meu
- mev
- mey
- mez
- mfa
- mfb
- mfc
- mfd
- mfe
- mfg
- mfh
- mfi
- mfj
- mfk
- mfl
- mfm
- mfn
- mfo
- mfq
- mfv
- mfy
- mfz
- mgb
- mgc
- mgd
- mgf
- mgg
- mgh
- mgi
- mgk
- mgl
- mgm
- mgo
- mgp
- mgr
- mgu
- mgw
- mhc
- mhi
- mhk
- mhl
- mho
- mhp
- mhs
- mhu
- mhw
- mhx
- mhy
- mhz
- mib
- mic
- mie
- mif
- mig
- mih
- mii
- mij
- mil
- mim
- min
- mio
- mip
- miq
- mir
- mit
- miu
- mix
- miy
- miz
- mjc
- mjg
- mji
- mjl
- mjs
- mjt
- mjv
- mjw
- mjx
- mjz
- mkb
- mkc
- mkd
- mke
- mkf
- mkg
- mki
- mkk
- mkl
- mkn
- mks
- mku
- mkw
- mkz
- mla
- mle
- mlf
- mlk
- mlm
- mln
- mlq
- mls
- mlt
- mlu
- mlv
- mlw
- mlx
- mma
- mmc
- mmd
- mme
- mmg
- mmh
- mml
- mmm
- mmn
- mmp
- mmx
- mmy
- mmz
- mnb
- mne
- mnf
- mng
- mni
- mnj
- mnk
- mnl
- mnm
- mnp
- mnu
- mnv
- mnw
- mnx
- mnz
- moa
- moc
- moe
- mog
- moh
- moi
- moj
- mop
- mor
- mos
- mot
- mov
- mox
- moy
- moz
- mpc
- mpd
- mpe
- mpg
- mph
- mpj
- mpm
- mpn
- mpq
- mpr
- mps
- mpt
- mpx
- mqg
- mqh
- mqj
- mql
- mqn
- mqu
- mqx
- mqz
- mrd
- mrf
- mrg
- mrh
- mri
- mrl
- mrm
- mrn
- mro
- mrp
- mrq
- mrr
- mrt
- mrw
- mrz
- msc
- mse
- msg
- msh
- msi
- msj
- msk
- msl
- msm
- msn
- msw
- msy
- mta
- mtb
- mtd
- mte
- mtf
- mtg
- mti
- mtk
- mtl
- mto
- mtp
- mtq
- mtr
- mtt
- mtu
- mua
- mug
- muh
- mui
- muk
- mum
- muo
- mup
- mur
- mus
- mut
- muv
- muy
- muz
- mva
- mve
- mvf
- mvg
- mvn
- mvo
- mvp
- mvv
- mvz
- mwa
- mwc
- mwe
- mwf
- mwg
- mwi
- mwm
- mwn
- mwp
- mwq
- mwt
- mwv
- mww
- mxa
- mxb
- mxd
- mxe
- mxh
- mxj
- mxl
- mxm
- mxn
- mxp
- mxq
- mxs
- mxt
- mxu
- mxv
- mxx
- mxy
- mya
- myb
- mye
- myh
- myk
- myl
- mym
- myp
- myu
- myw
- myx
- myy
- mza
- mzb
- mzi
- mzj
- mzk
- mzl
- mzm
- mzn
- mzp
- mzq
- mzr
- mzv
- mzw
- mzz
- nab
- nac
- nag
- naj
- nak
- nal
- nan
- nao
- nap
- naq
- nar
- nas
- nat
- nau
- nav
- naw
- naz
- nba
- nbb
- nbc
- nbe
- nbh
- nbi
- nbl
- nbm
- nbn
- nbp
- nbq
- nbr
- nbu
- nbv
- ncb
- nce
- ncf
- ncg
- ncj
- ncl
- ncm
- nco
- ncq
- ncr
- ncu
- nda
- ndb
- ndc
- ndd
- nde
- ndh
- ndi
- ndm
- ndo
- ndp
- ndr
- nds
- ndu
- ndv
- ndx
- ndy
- ndz
- neb
- nen
- neq
- ner
- nes
- nev
- new
- ney
- nez
- nfa
- nfd
- nfl
- nfr
- nfu
- nga
- ngb
- ngc
- nge
- ngi
- ngj
- ngl
- ngn
- ngs
- ngt
- ngu
- ngw
- ngz
- nhb
- nhd
- nhe
- nhg
- nhi
- nhn
- nhp
- nhr
- nhu
- nhv
- nhw
- nhx
- nhy
- nhz
- nia
- nid
- nih
- nii
- nij
- nil
- nim
- nin
- niq
- nir
- nit
- niu
- niw
- nix
- niy
- niz
- nja
- njb
- njh
- njj
- njm
- njn
- njo
- njs
- njx
- njz
- nka
- nkb
- nke
- nkh
- nkk
- nko
- nku
- nkw
- nkx
- nlc
- nld
- nlg
- nli
- nlj
- nlk
- nlo
- nlu
- nlv
- nlx
- nma
- nmb
- nmc
- nmf
- nmh
- nmi
- nmk
- nmm
- nmn
- nmo
- nms
- nmz
- nna
- nnb
- nnc
- nnd
- nng
- nni
- nnj
- nnm
- nno
- nnp
- nnu
- nnw
- nnz
- noa
- nod
- noe
- nof
- nos
- not
- nou
- noz
- npb
- nph
- npi
- npl
- nps
- npy
- nqg
- nqt
- nqy
- nre
- nrf
- nrg
- nri
- nsa
- nsm
- nso
- nst
- nti
- ntj
- ntk
- ntm
- nto
- ntp
- ntr
- ntu
- nud
- nuf
- nuj
- nuk
- nun
- nuo
- nup
- nuq
- nus
- nut
- nux
- nuy
- nwb
- nwi
- nwm
- nxa
- nxd
- nxg
- nxk
- nxq
- nxr
- nya
- nyb
- nyd
- nyf
- nyg
- nyh
- nyi
- nyj
- nyk
- nym
- nyn
- nyo
- nyq
- nys
- nyu
- nyw
- nyy
- nza
- nzb
- nzi
- nzk
- nzm
- nzy
- obo
- ocu
- odk
- odu
- ofu
- ogb
- ogc
- ogg
- ogo
- oia
- ojb
- oka
- oke
- okh
- oki
- okr
- oks
- oku
- okv
- okx
- ola
- old
- olu
- oma
- omb
- one
- ong
- oni
- onj
- onn
- ono
- onp
- ont
- ood
- opa
- opm
- ora
- orc
- ore
- org
- orh
- oro
- ors
- ort
- oru
- orx
- ory
- orz
- osi
- oso
- oss
- ost
- otd
- ote
- otm
- otq
- otr
- ots
- ott
- otx
- oub
- owi
- oyb
- oyd
- oym
- ozm
- pab
- pac
- pad
- pag
- pah
- pai
- pak
- pam
- pan
- pao
- pap
- pau
- pav
- pay
- pbb
- pbc
- pbg
- pbi
- pbl
- pbm
- pbn
- pbo
- pbp
- pbs
- pbt
- pbu
- pbv
- pca
- pcb
- pcc
- pce
- pcf
- pcg
- pch
- pci
- pcj
- pck
- pcl
- pcm
- pcn
- pcw
- pdc
- pdn
- pdo
- pdt
- pdu
- peb
- peg
- pei
- pek
- pem
- pes
- pex
- pfe
- pga
- pgg
- pha
- phk
- phl
- phq
- phr
- pht
- pia
- pib
- pic
- pid
- pih
- pil
- pio
- pip
- pir
- pis
- piu
- piv
- piy
- pjt
- pkb
- pkg
- pkh
- pko
- pkt
- pku
- plc
- plg
- plj
- plk
- pll
- pln
- plr
- pls
- plt
- plu
- plv
- plw
- pma
- pmf
- pmi
- pmj
- pmm
- pmq
- pmx
- pmy
- pnb
- pnc
- pne
- png
- pnq
- pnu
- pny
- pnz
- poc
- poe
- pof
- poh
- poi
- pol
- pon
- poo
- por
- pos
- pot
- pov
- pow
- poy
- ppi
- ppk
- ppl
- ppm
- ppo
- ppq
- ppt
- pqa
- pqm
- prc
- prf
- pri
- prm
- prn
- prs
- prt
- pru
- prx
- psa
- pse
- psh
- psi
- psn
- pss
- pst
- psw
- pta
- ptu
- pua
- puc
- pud
- pug
- pui
- pum
- puo
- puu
- pwa
- pwb
- pwg
- pwm
- pwn
- pwo
- pwr
- pww
- pxm
- pym
- pyu
- qub
- quc
- qud
- qug
- quh
- qui
- qul
- qum
- qun
- qus
- quv
- quw
- qux
- quy
- quz
- qvi
- qvj
- qvm
- qvn
- qvo
- qvs
- qvw
- qwa
- qwh
- qws
- qxa
- qxl
- qxn
- qxp
- qxq
- qxs
- qxu
- raa
- rab
- rad
- raf
- rag
- rah
- rai
- ral
- ram
- rao
- rar
- rat
- rau
- rav
- raw
- rbb
- rcf
- rdb
- rei
- rej
- rel
- res
- rey
- rgs
- rgu
- rhg
- rhp
- ria
- rif
- ril
- rim
- rin
- rir
- rji
- rjs
- rki
- rkm
- rmb
- rmc
- rml
- rmn
- rmo
- rmq
- rmt
- rmy
- rmz
- rnd
- rnl
- rog
- roh
- rol
- ron
- roo
- row
- rro
- rsw
- rtm
- rue
- ruf
- rug
- rui
- ruk
- run
- rus
- ruy
- ruz
- rwa
- rwk
- rwo
- rwr
- ryu
- saa
- sab
- sac
- sad
- saf
- sag
- sah
- saj
- san
- sao
- saq
- sas
- sat
- sau
- sav
- saw
- sax
- say
- saz
- sba
- sbb
- sbc
- sbd
- sbe
- sbg
- sbh
- sbk
- sbl
- sbn
- sbp
- sbr
- sbs
- sbu
- sbx
- sby
- sbz
- sce
- scg
- sch
- sck
- scl
- scn
- scp
- scs
- sct
- scu
- scv
- scw
- sda
- sde
- sdg
- sdh
- sdo
- sdp
- sdq
- sdr
- sea
- sed
- see
- sef
- seg
- seh
- sei
- sek
- sen
- sep
- ses
- set
- sev
- sew
- sey
- sez
- sfm
- sfw
- sgb
- sgc
- sgd
- sge
- sgh
- sgi
- sgj
- sgp
- sgr
- sgw
- sgy
- sgz
- sha
- shb
- shc
- she
- shg
- shh
- shi
- shj
- shk
- shm
- shn
- sho
- shp
- shq
- shr
- shs
- shu
- shw
- shy
- sid
- sie
- sif
- sig
- sil
- sin
- sip
- sir
- siu
- siw
- siy
- sja
- sjb
- sjg
- sjl
- sjm
- sjo
- sjp
- sjr
- skb
- skd
- skg
- skj
- skn
- skq
- skr
- skt
- sku
- skv
- skx
- sky
- slc
- sld
- sle
- slk
- slp
- slr
- slu
- slv
- slx
- sly
- slz
- sme
- smf
- smh
- sml
- smn
- smo
- smq
- smt
- smu
- smw
- smy
- sna
- snc
- snd
- sne
- snf
- sng
- snk
- snl
- snm
- snn
- snp
- snq
- sns
- snv
- snw
- sny
- soa
- sob
- soc
- soe
- soi
- sok
- sol
- som
- soo
- sop
- soq
- sor
- sos
- sot
- sou
- soy
- soz
- spa
- spm
- spn
- spo
- spp
- sps
- spt
- spu
- spy
- sqq
- srb
- src
- sre
- srl
- srm
- srn
- sro
- srp
- srq
- srr
- sru
- srx
- sry
- srz
- ssb
- sse
- ssi
- ssk
- ssn
- sso
- sss
- sst
- ssw
- ssx
- ssy
- stf
- sti
- stj
- stk
- sto
- stp
- sts
- stt
- stv
- sua
- sug
- sui
- suj
- suk
- sun
- suq
- sur
- sus
- suv
- suy
- suz
- svb
- svs
- swb
- swc
- swe
- swh
- swi
- swj
- swk
- swo
- swp
- swr
- swv
- sxb
- sxn
- sxw
- sya
- syb
- syk
- syl
- sym
- sys
- syw
- szb
- szg
- szp
- szv
- tab
- tac
- tah
- taj
- tak
- tal
- tam
- tan
- tao
- tap
- taq
- tar
- tat
- tau
- tav
- taw
- tay
- taz
- tba
- tbc
- tbf
- tbg
- tbj
- tbk
- tbl
- tbo
- tbp
- tbt
- tby
- tbz
- tca
- tcc
- tcd
- tce
- tcf
- tcn
- tcp
- tcs
- tcu
- tcx
- tcy
- tcz
- tdb
- tdc
- tdd
- tdf
- tdg
- tdh
- tdj
- tdk
- tdl
- tdn
- tdo
- tds
- tdt
- tdv
- tdx
- tdy
- ted
- tee
- tef
- tei
- tek
- tel
- tem
- teo
- teq
- ter
- tes
- tet
- tew
- tex
- tfi
- tfn
- tfr
- tft
- tga
- tgc
- tgd
- tgj
- tgk
- tgl
- tgo
- tgp
- tgs
- tgw
- tgy
- tha
- thd
- the
- thf
- thk
- thl
- thm
- thp
- thq
- thr
- ths
- thy
- thz
- tic
- tif
- tig
- tii
- tik
- tio
- tiq
- tir
- tis
- tiv
- tiw
- tix
- tiy
- tja
- tjg
- tji
- tkb
- tkd
- tke
- tkg
- tkp
- tkq
- tkt
- tku
- tkx
- tla
- tlb
- tld
- tlf
- tli
- tlj
- tll
- tlp
- tlq
- tlr
- tls
- tlx
- tma
- tmc
- tmd
- tmf
- tml
- tmn
- tmq
- tmy
- tna
- tnb
- tnc
- tng
- tnk
- tnl
- tnm
- tnn
- tnp
- tnr
- tnt
- tnv
- tny
- tob
- toc
- tod
- tof
- tog
- toj
- tol
- tom
- ton
- too
- top
- toq
- tos
- tou
- tov
- tow
- tpa
- tpe
- tpi
- tpj
- tpl
- tpm
- tpp
- tpq
- tpr
- tpu
- tpx
- tqo
- tqu
- tra
- trc
- trd
- trf
- tri
- trn
- tro
- trp
- trq
- trs
- tru
- trv
- tsa
- tsb
- tsc
- tsg
- tsi
- tsj
- tsn
- tso
- tsp
- tsr
- tsu
- tsv
- tsw
- tsx
- tsz
- ttb
- ttc
- tte
- tth
- tti
- ttj
- ttk
- ttm
- tto
- ttq
- ttr
- tts
- ttv
- ttw
- tty
- tuc
- tue
- tuf
- tug
- tui
- tuk
- tul
- tum
- tuo
- tuq
- tur
- tus
- tuv
- tuy
- tuz
- tva
- tvd
- tvk
- tvl
- tvn
- tvs
- tvt
- tvu
- twb
- twe
- twf
- twh
- twi
- twm
- twp
- twr
- twu
- tww
- twx
- twy
- txa
- txn
- txo
- txq
- txt
- txu
- txy
- tye
- tyn
- tyr
- tyv
- tyy
- tyz
- tzh
- tzj
- tzm
- tzo
- uar
- uba
- ubr
- ubu
- udg
- udl
- udu
- ugo
- uhn
- uig
- uis
- uiv
- uki
- ukp
- ukr
- ukw
- ula
- ulu
- umb
- umm
- ums
- umu
- une
- ung
- unr
- unx
- upv
- ura
- urb
- urd
- urh
- uri
- urk
- url
- urt
- ury
- usa
- usi
- usp
- uss
- uta
- ute
- uth
- utr
- uuu
- uya
- uzn
- uzs
- vaa
- vaf
- vag
- vah
- vai
- vaj
- vam
- van
- vap
- var
- vas
- vav
- vay
- vem
- ven
- ver
- vie
- vif
- vig
- viv
- vkl
- vkn
- vls
- vmc
- vmh
- vmj
- vmk
- vmm
- vmp
- vmw
- vmx
- vmz
- vnk
- vor
- vra
- vrs
- vum
- vun
- vut
- wad
- wal
- wan
- wap
- war
- was
- wat
- wau
- waw
- way
- wbb
- wbf
- wbi
- wbj
- wbk
- wbl
- wbm
- wbp
- wbq
- wbr
- wca
- wci
- wdd
- wdj
- wed
- weh
- wem
- weo
- wes
- wew
- wgb
- wgi
- whg
- wib
- wic
- wim
- win
- wiu
- wja
- wji
- wkd
- wlc
- wle
- wli
- wlo
- wlv
- wlw
- wlx
- wmb
- wmd
- wme
- wmo
- wms
- wmt
- wmw
- wnc
- wni
- wno
- wnp
- wob
- wod
- wof
- wog
- wol
- wom
- won
- wow
- wrk
- wrm
- wro
- wrp
- wrs
- wry
- wsa
- wsi
- wsk
- wss
- wti
- wtm
- wud
- wut
- wuu
- wuv
- wwa
- wwo
- wyy
- xac
- xal
- xav
- xbi
- xbr
- xdo
- xdy
- xed
- xem
- xer
- xes
- xgu
- xho
- xkb
- xkc
- xkf
- xkg
- xkj
- xkk
- xkl
- xkn
- xks
- xkt
- xkv
- xky
- xkz
- xla
- xmc
- xmf
- xmg
- xmh
- xmm
- xmt
- xmv
- xmw
- xmz
- xnr
- xns
- xnz
- xod
- xog
- xok
- xom
- xon
- xpe
- xra
- xrb
- xri
- xrw
- xsb
- xsm
- xsn
- xsq
- xsr
- xsu
- xta
- xtc
- xtd
- xte
- xti
- xtj
- xtl
- xtm
- xtn
- xtt
- xty
- xub
- xuj
- xuu
- xvi
- xwe
- xwg
- xwl
- yaa
- yad
- yae
- yaf
- yah
- yak
- yal
- yam
- yan
- yao
- yap
- yaq
- yaw
- yax
- yay
- yaz
- yba
- ybb
- ybe
- ybh
- ybi
- ybj
- ybl
- ycl
- ycn
- ydd
- yde
- ydg
- yea
- yer
- 'yes'
- yet
- yeu
- yev
- yey
- ygr
- ygw
- yhd
- yif
- yig
- yij
- yim
- yin
- yiq
- yis
- yiu
- yix
- yiz
- yka
- ykg
- yki
- ykk
- ykm
- yle
- yll
- ymb
- ymk
- yml
- ymm
- yno
- ynq
- yns
- yog
- yom
- yon
- yor
- yot
- yoy
- yra
- yrb
- yre
- yrl
- ysn
- ysp
- yss
- yua
- yue
- yuf
- yui
- yuj
- yum
- yun
- yup
- yuq
- yur
- yuy
- yuz
- yva
- ywa
- ywl
- ywn
- ywq
- yyu
- zaa
- zab
- zac
- zad
- zae
- zaf
- zag
- zai
- zaj
- zak
- zam
- zao
- zap
- zar
- zas
- zat
- zau
- zav
- zaw
- zay
- zaz
- zbc
- zbu
- zca
- zcd
- zdj
- zeh
- zem
- zgb
- zhi
- zia
- zik
- zim
- zin
- ziw
- zkd
- zkn
- zkr
- zlj
- zlm
- zln
- zmb
- zmp
- zmq
- zms
- zne
- zng
- zns
- zoc
- zoh
- zom
- zos
- zpa
- zpc
- zpd
- zpe
- zpg
- zph
- zpj
- zpk
- zpl
- zpm
- zpn
- zpo
- zpp
- zpq
- zpr
- zps
- zpu
- zpv
- zpw
- zpx
- zpy
- zpz
- zrg
- zro
- zrs
- zsm
- zte
- ztg
- ztl
- ztp
- ztq
- zts
- ztx
- zty
- zua
- zul
- zun
- zuy
- zwa
- zyb
- zyg
- zyj
- zyn
- zyp
- zzj
XEUS - A Cross-lingual Encoder for Universal Speech
XEUS is a large-scale multilingual speech encoder by Carnegie Mellon University's WAVLab that covers over 4000 languages. It is pre-trained on over 1 million hours of publicly available speech datasets. It requires fine-tuning to be used in downstream tasks such as Speech Recognition or Translation. Its hidden states can also be used with k-means for semantic Speech Tokenization. XEUS uses the E-Branchformer architecture and is trained using HuBERT-style masked prediction of discrete speech tokens extracted from WavLabLM. During training, the input speech is also augmented with acoustic noise and reverberation, making XEUS more robust. The total model size is 577M parameters.
XEUS tops the ML-SUPERB multilingual speech recognition leaderboard, outperforming MMS, w2v-BERT 2.0, and XLS-R. XEUS also sets a new state-of-the-art on 4 tasks in the monolingual SUPERB benchmark.
More information about XEUS, including download links for our crawled 4000-language dataset, can be found in the project page and paper.
Requirements
The code for XEUS is still in progress of being merged into the main ESPnet repo. It can instead be used from the following fork:
pip install 'espnet @ git+https://github.com/wanchichen/espnet.git@ssl'
git lfs install
git clone https://huggingface.co/espnet/XEUS
XEUS supports Flash Attention, which can be installed as follows:
pip install flash-attn --no-build-isolation
Usage
from torch.nn.utils.rnn import pad_sequence
from espnet2.tasks.ssl import SSLTask
import soundfile as sf
device = "cuda" if torch.cuda.is_available() else "cpu"
xeus_model, xeus_train_args = SSLTask.build_model_from_file(
None,
'/path/to/checkpoint/here/checkpoint.pth',
device,
)
wavs, sampling_rate = sf.read('/path/to/audio.wav') # sampling rate should be 16000
wav_lengths = torch.LongTensor([len(wav) for wav in [wavs]]).to(device)
wavs = pad_sequence(torch.Tensor([wavs]), batch_first=True).to(device)
# we recommend use_mask=True during fine-tuning
feats = xeus_model.encode(wavs, wav_lengths, use_mask=False, use_final_output=False)[0][-1] # take the output of the last layer -> batch_size x seq_len x hdim
With Flash Attention:
[layer.use_flash_attn = True for layer in xeus_model.encoder.encoders]
with torch.cuda.amp.autocast(dtype=torch.bfloat16):
feats = xeus_model.encode(wavs, wav_lengths, use_mask=False, use_final_output=False)[0][-1]
Tune the masking settings:
xeus_model.masker.mask_prob = 0.65 # default 0.8
xeus_model.masker.mask_length = 20 # default 10
xeus_model.masker.mask_selection = 'static' # default 'uniform'
xeus_model.train()
feats = xeus_model.encode(wavs, wav_lengths, use_mask=True, use_final_output=False)[0][-1]
Results
@misc{chen2024robustspeechrepresentationlearning,
title={Towards Robust Speech Representation Learning for Thousands of Languages},
author={William Chen and Wangyou Zhang and Yifan Peng and Xinjian Li and Jinchuan Tian and Jiatong Shi and Xuankai Chang and Soumi Maiti and Karen Livescu and Shinji Watanabe},
year={2024},
eprint={2407.00837},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2407.00837},
}