Datasets:

acojaco
/

cmudict-espeak-ipa

word large_stringlengths 1 28	raw_word large_stringlengths 1 28	ipa large_stringlengths 1 35	espeak_ascii large_stringlengths 1 35	arpabet large_stringlengths 1 80	stress_pattern large_stringclasses 243 values	syllables int64 0 12	variant_index int64 0 3
'bout	'bout	bˈaʊt	b'aUt	B AW1 T	1	1	0
'cause	'cause	kəz	k@z	K AH0 Z	0	1	0
'course	'course	kˈɔːɹs	k'O:rs	K AO1 R S	1	1	0
'cuse	'cuse	kjˈuːz	kj'u:z	K Y UW1 Z	1	1	0
'em	'em	əm	@m	AH0 M	0	1	0
'frisco	'frisco	fɹˈɪskoʊ	fr'IskoU	F R IH1 S K OW0	1-0	2	0
'gain	'gain	ɡˈɛn	g'En	G EH1 N	1	1	0
'kay	'kay	kˈeɪ	k'eI	K EY1	1	1	0
'm	'm	əm	@m	AH0 M	0	1	0
'n	'n	ən	@n	AH0 N	0	1	0
'round	'round	ɹˈaʊnd	r'aUnd	R AW1 N D	1	1	0
's	's	ˈɛs	'Es	EH1 S	1	1	0
'til	'til	tˈɪl	t'Il	T IH1 L	1	1	0
'tis	'tis	tˈɪz	t'Iz	T IH1 Z	1	1	0
'twas	'twas	twˈʌz	tw'Vz	T W AH1 Z	1	1	0
a	a	ə	@	AH0	0	1	0
a	a(2)	ˈeɪ	'eI	EY1	1	1	1
a's	a's	ˈeɪz	'eIz	EY1 Z	1	1	0
a.	a.	ˈeɪ	'eI	EY1	1	1	0
a.'s	a.'s	ˈeɪz	'eIz	EY1 Z	1	1	0
a.d.	a.d.	ˌeɪdˈiː	,eId'i:	EY2 D IY1	2-1	2	0
a.m.	a.m.	ˌeɪˈɛm	,eI'Em	EY2 EH1 M	2-1	2	0
a.s	a.s	ˈeɪz	'eIz	EY1 Z	1	1	0
aaa	aaa	tɹˌɪpəlˈeɪ	tr,Ip@l'eI	T R IH2 P AH0 L EY1	2-0-1	3	0
aaberg	aaberg	ˈɑːbɚɡ	'A:b3g	AA1 B ER0 G	1-0	2	0
aachen	aachen	ˈɑːkən	'A:k@n	AA1 K AH0 N	1-0	2	0
aachener	aachener	ˈɑːkənɚ	'A:k@n3	AA1 K AH0 N ER0	1-0-0	3	0
aaker	aaker	ˈɑːkɚ	'A:k3	AA1 K ER0	1-0	2	0
aalborg	aalborg	ˈɔːlbɔːɹɡ	'O:lbO:rg	AO1 L B AO0 R G # place, danish	1-0	2	0
aalborg	aalborg(2)	ˈɑːlbɔːɹɡ	'A:lbO:rg	AA1 L B AO0 R G	1-0	2	1
aalburg	aalburg	ˈælbɚɡ	'alb3g	AE1 L B ER0 G # place, dutch	1-0	2	0
aalen	aalen	ˈælən	'al@n	AE1 L AH0 N # place, german	1-0	2	0
aalen	aalen(2)	ˈɑːlən	'A:l@n	AA1 L AH0 N	1-0	2	1
aaliyah	aaliyah	ˌɑːlˈiːˌɑː	,A:l'i:,A:	AA2 L IY1 AA2	2-1-2	3	0
aalseth	aalseth	ˈɑːlsɛθ	'A:lsET	AA1 L S EH0 TH	1-0	2	0
aalsmeer	aalsmeer	ˈɑːlsmɪɹ	'A:lsmIr	AA1 L S M IH0 R # place, dutch	1-0	2	0
aalto	aalto	ˈɑːltˌoʊ	'A:lt,oU	AA1 L T OW2 # name, finnish	1-2	2	0
aamodt	aamodt	ˈɑːmət	'A:m@t	AA1 M AH0 T	1-0	2	0
aancor	aancor	ˈɑːnkˌɔːɹ	'A:nk,O:r	AA1 N K AO2 R	1-2	2	0
aardema	aardema	ɑːɹdˈɛmə	A:rd'Em@	AA0 R D EH1 M AH0	0-1-0	3	0
aardvark	aardvark	ˈɑːɹdvˌɑːɹk	'A:rdv,A:rk	AA1 R D V AA2 R K	1-2	2	0
aardvarks	aardvarks	ˈɑːɹdvˌɑːɹks	'A:rdv,A:rks	AA1 R D V AA2 R K S	1-2	2	0
aargh	aargh	ˈɑːɹɡ	'A:rg	AA1 R G	1	1	0
aarhus	aarhus	ˌɑːhˈuːs	,A:h'u:s	AA2 HH UW1 S	2-1	2	0
aaron	aaron	ˈɛɹən	'Er@n	EH1 R AH0 N	1-0	2	0
aaron's	aaron's	ˈɛɹənz	'Er@nz	EH1 R AH0 N Z	1-0	2	0
aarons	aarons	ˈɛɹənz	'Er@nz	EH1 R AH0 N Z	1-0	2	0
aaronson	aaronson	ˈɛɹənsən	'Er@ns@n	EH1 R AH0 N S AH0 N	1-0-0	3	0
aaronson	aaronson(2)	ˈɑːɹənsən	'A:r@ns@n	AA1 R AH0 N S AH0 N	1-0-0	3	1
aaronson's	aaronson's	ˈɛɹənsənz	'Er@ns@nz	EH1 R AH0 N S AH0 N Z	1-0-0	3	0
aaronson's	aaronson's(2)	ˈɑːɹənsənz	'A:r@ns@nz	AA1 R AH0 N S AH0 N Z	1-0-0	3	1
aarti	aarti	ˈɑːɹtˌiː	'A:rt,i:	AA1 R T IY2	1-2	2	0
aase	aase	ˈɑːs	'A:s	AA1 S	1	1	0
aasen	aasen	ˈɑːsən	'A:s@n	AA1 S AH0 N	1-0	2	0
ab	ab	ˈæb	'ab	AE1 B	1	1	0
ab	ab(2)	ˈeɪbˌiː	'eIb,i:	EY1 B IY1	1-2	2	1
aba	aba	ˌeɪbˌiːˈeɪ	,eIb,i:'eI	EY2 B IY2 EY1	2-2-1	3	0
ababa	ababa	əbˈɑːbə	@b'A:b@	AH0 B AA1 B AH0	0-1-0	3	0
ababa	ababa(2)	ˈɑːbəbə	'A:b@b@	AA1 B AH0 B AH0	1-0-0	3	1
abacha	abacha	ˈæbəkə	'ab@k@	AE1 B AH0 K AH0	1-0-0	3	0
aback	aback	əbˈæk	@b'ak	AH0 B AE1 K	0-1	2	0
abaco	abaco	ˈæbəkˌoʊ	'ab@k,oU	AE1 B AH0 K OW2	1-0-2	3	0
abacus	abacus	ˈæbəkəs	'ab@k@s	AE1 B AH0 K AH0 S	1-0-0	3	0
abad	abad	əbˈɑːd	@b'A:d	AH0 B AA1 D	0-1	2	0
abadaka	abadaka	əbˈædəkə	@b'ad@k@	AH0 B AE1 D AH0 K AH0	0-1-0-0	4	0
abadi	abadi	əbˈædiː	@b'adi:	AH0 B AE1 D IY0	0-1-0	3	0
abadie	abadie	əbˈædiː	@b'adi:	AH0 B AE1 D IY0	0-1-0	3	0
abair	abair	əbˈɛɹ	@b'Er	AH0 B EH1 R	0-1	2	0
abalkin	abalkin	əbˈɑːlkɪn	@b'A:lkIn	AH0 B AA1 L K IH0 N	0-1-0	3	0
abalone	abalone	ˌæbəlˈoʊniː	,ab@l'oUni:	AE2 B AH0 L OW1 N IY0	2-0-1-0	4	0
abalones	abalones	ˌæbəlˈoʊniːz	,ab@l'oUni:z	AE2 B AH0 L OW1 N IY0 Z	2-0-1-0	4	0
abalos	abalos	ɑːbˈɑːloʊz	A:b'A:loUz	AA0 B AA1 L OW0 Z	0-1-0	3	0
abandon	abandon	əbˈændən	@b'and@n	AH0 B AE1 N D AH0 N	0-1-0	3	0
abandoned	abandoned	əbˈændənd	@b'and@nd	AH0 B AE1 N D AH0 N D	0-1-0	3	0
abandoning	abandoning	əbˈændənɪŋ	@b'and@nIN	AH0 B AE1 N D AH0 N IH0 NG	0-1-0-0	4	0
abandonment	abandonment	əbˈændənmənt	@b'and@nm@nt	AH0 B AE1 N D AH0 N M AH0 N T	0-1-0-0	4	0
abandonments	abandonments	əbˈændənmənts	@b'and@nm@nts	AH0 B AE1 N D AH0 N M AH0 N T S	0-1-0-0	4	0
abandons	abandons	əbˈændənz	@b'and@nz	AH0 B AE1 N D AH0 N Z	0-1-0	3	0
abanto	abanto	əbˈæntoʊ	@b'antoU	AH0 B AE1 N T OW0	0-1-0	3	0
abarca	abarca	əbˈɑːɹkə	@b'A:rk@	AH0 B AA1 R K AH0	0-1-0	3	0
abare	abare	ɑːbˈɑːɹiː	A:b'A:ri:	AA0 B AA1 R IY0	0-1-0	3	0
abascal	abascal	ˈæbəskəl	'ab@sk@l	AE1 B AH0 S K AH0 L	1-0-0	3	0
abash	abash	əbˈæʃ	@b'aS	AH0 B AE1 SH	0-1	2	0
abashed	abashed	əbˈæʃt	@b'aSt	AH0 B AE1 SH T	0-1	2	0
abasia	abasia	əbˈeɪʒjə	@b'eIZj@	AH0 B EY1 ZH Y AH0	0-1-0	3	0
abate	abate	əbˈeɪt	@b'eIt	AH0 B EY1 T	0-1	2	0
abated	abated	əbˈeɪtɪd	@b'eItId	AH0 B EY1 T IH0 D	0-1-0	3	0
abatement	abatement	əbˈeɪtmənt	@b'eItm@nt	AH0 B EY1 T M AH0 N T	0-1-0	3	0
abatements	abatements	əbˈeɪtmənts	@b'eItm@nts	AH0 B EY1 T M AH0 N T S	0-1-0	3	0
abates	abates	əbˈeɪts	@b'eIts	AH0 B EY1 T S	0-1	2	0
abating	abating	əbˈeɪtɪŋ	@b'eItIN	AH0 B EY1 T IH0 NG	0-1-0	3	0
abattoir	abattoir	ˌæbətwˈɑːɹ	,ab@tw'A:r	AE2 B AH0 T W AA1 R	2-0-1	3	0
abba	abba	ˈæbə	'ab@	AE1 B AH0	1-0	2	0
abbado	abbado	əbˈɑːdoʊ	@b'A:doU	AH0 B AA1 D OW0	0-1-0	3	0
abbas	abbas	əbˈɑːs	@b'A:s	AH0 B AA1 S	0-1	2	0
abbasi	abbasi	ɑːbˈɑːsiː	A:b'A:si:	AA0 B AA1 S IY0	0-1-0	3	0
abbate	abbate	ˈɑːbeɪt	'A:beIt	AA1 B EY0 T	1-0	2	0
abbatiello	abbatiello	ɑːbɑːtiːˈɛloʊ	A:bA:ti:'EloU	AA0 B AA0 T IY0 EH1 L OW0	0-0-0-1-0	5	0
abbe	abbe	ˈæbiː	'abi:	AE1 B IY0	1-0	2	0
abbe	abbe(2)	æbˈeɪ	ab'eI	AE0 B EY1	0-1	2	1

End of preview. Expand in Data Studio

CMUDict eSpeak-NG Aligned IPA Lexicon (130k+ words)

A standardized English pronunciation dataset derived from CMUdict (v0.7b), precompiled into eSpeak-NG compatible International Phonetic Alphabet (IPA) representation.

Dataset Highlights

eSpeak-NG style Dialect Alignment: Direct support for vowel duration markers (ː), rhotic reductions (ɚ, ɝ), and unreduced vowels (ʌ, ə). Prepared with ascii strings directly accepted as espeak_NG synthesis input.
Standardized Stress Positioning: Primary (ˈ) and secondary (ˌ) stress marks placed directly before the syllable nucleus.
Fast Lookup Table: Designed as a drop-in $O(1)$ dictionary for client-side (WebAssembly/WASM) and edge Neural TTS models (Piper, VITS, Kokoro, Matcha-TTS).

Dataset Schema

Column	Type	Description	Example
`word`	string	Normalized lowercase word token	`evaluate`
`raw_word`	string	Original token from CMUdict (with variants)	`evaluate(2)`
`ipa`	string	phonetic IPA sequence	`iːvˈæljuːˌeɪt`
`espeak_ascii`	string	IPA string in eSpeak-NG input form	`i:v'alju:,eIt`
`arpabet`	string	Original space-separated ARPAbet phones	`IY0 V AE1 L Y UW0 EY2 T`
`stress_pattern`	string	Numerical stress pattern per vowel nucleus	`0-1-0-2`
`syllables`	int32	Total syllable count	`4`
`variant_index`	int32	`0` for primary pronunciation, `>0` for alternative variants	`1`

Total Entries

135,166 phonetic entries.

Upstream License

Downloads last month: 58