Datasets:

AfriSpeech
/

gloss-mt-train-en

AfriSpeech/gloss-mt-train-en provides full-scale parallel datasets engineered specifically for glossary-constrained machine translation from English into African languages.

Each training example features nominal phrases in the source sentence replaced in-place with their verified target language translation enclosed in <term> markers (e.g. The <kannifoɔ> made a covenant with God...).

Author: Mich-Seth Owusu (michseth@ghananlp.org)
Organization: AfriSpeech & Ghana NLP
Codebase: github.com/AfriSpeech/afrigloss-mt
Source Language: English (en)
Total African Subsets: 723 languages
Total Parallel Sentences: 19,477,537

Supported Language Subsets

Subset (`config`)	Target Language	Train	Validation	Test	Total Examples
`aab`	AAB	5,249	291	291	5,831
`abn`	ABN	20,916	1,162	1,162	23,240
`abr`	ABR	5,863	325	325	6,513
`acd`	ACD	5,055	280	280	5,615
`ach`	Acholi	78,649	4,369	4,369	87,387
`ada`	Dangme / Ada	87,671	4,870	4,870	97,411
`adh`	ADH	21,774	1,209	1,209	24,192
`adj`	ADJ	5,321	295	295	5,911
`afo`	AFO	5,356	297	297	5,950
`afr`	AFR	189,360	10,519	10,519	210,398
`agq`	AGQ	5,297	294	294	5,885
`ags`	AGS	5,702	316	316	6,334
`aha`	AHA	25,807	1,433	1,433	28,673
`ajg`	AJG	40,063	2,225	2,225	44,513
`akp`	AKP	5,452	302	302	6,056
`ala`	ALA	5,558	308	308	6,174
`ald`	ALD	5,490	304	304	6,098
`alw`	ALW	3,340	185	185	3,710
`alz`	ALZ	23,171	1,287	1,287	25,745
`amh`	Amharic	221,557	12,308	12,308	246,173
`ann`	ANN	36,675	2,037	2,037	40,749
`anu`	ANU	19,411	1,078	1,078	21,567
`anv`	ANV	5,554	308	308	6,170
`any`	ANY	5,352	297	297	5,946
`apd`	APD	6,038	335	335	6,708
`ar`	African Arabic	177,609	9,867	9,867	197,343
`arq`	ARQ	5,290	293	293	5,876
`ary`	ARY	3,816	211	211	4,238
`asa`	ASA	6,099	338	338	6,775
`asg`	ASG	19,069	1,059	1,059	21,187
`asj`	ASJ	5,394	299	299	5,992
`atg`	ATG	5,735	318	318	6,371
`auh`	AUH	887	49	49	985
`avn`	AVN	5,833	324	324	6,481
`avu`	AVU	4,306	239	239	4,784
`azo`	AZO	5,145	285	285	5,715
`bag`	BAG	5,687	315	315	6,317
`bas`	BAS	57,259	3,181	3,181	63,621
`bav`	BAV	5,774	320	320	6,414
`bba`	BBA	17,910	995	995	19,900
`bbj`	BBJ	9,421	523	523	10,467
`bbk`	BBK	5,599	310	310	6,219
`bci`	BCI	73,375	4,076	4,076	81,527
`bcn`	BCN	5,198	288	288	5,774
`bcv`	BCV	5,078	282	282	5,642
`bcw`	BCW	5,454	302	302	6,058
`bcy`	BCY	5,266	292	292	5,850
`bdh`	BDH	6,388	354	354	7,096
`bds`	BDS	6,042	335	335	6,712
`bej`	BEJ	4,682	260	260	5,202
`bem`	Bemba	141,685	7,871	7,871	157,427
`beq`	BEQ	27,757	1,541	1,541	30,839
`bet`	BET	4,417	245	245	4,907
`bev`	BEV	4,824	267	267	5,358
`bex`	BEX	7,177	398	398	7,973
`bez`	BEZ	6,774	376	376	7,526
`bfa`	BFA	23,400	1,299	1,299	25,998
`bfd`	BFD	5,142	285	285	5,712
`bfm`	BFM	5,545	307	307	6,159
`bfo`	BFO	5,946	330	330	6,606
`bhs`	BHS	5,875	326	326	6,527
`bib`	BIB	5,882	326	326	6,534
`bim`	BIM	22,772	1,265	1,265	25,302
`bin`	BIN	64,801	3,600	3,600	72,001
`biv`	BIV	5,436	302	302	6,040
`bjv`	BJV	8,243	457	457	9,157
`bjw`	BJW	6,609	367	367	7,343
`bkh`	BKH	5,624	312	312	6,248
`bkv`	BKV	11,066	614	614	12,294
`bkw`	BKW	5,972	331	331	6,634
`bky`	BKY	20,582	1,143	1,143	22,868
`bm`	BM	24,619	1,367	1,367	27,353
`bma`	BMA	4,220	234	234	4,688
`bmo`	BMO	5,976	332	332	6,640
`bmv`	BMV	5,430	301	301	6,032
`bng`	BNG	13,759	764	764	15,287
`bni`	BNI	8,566	475	475	9,516
`bom`	BOM	22,268	1,237	1,237	24,742
`bou`	BOU	5,635	313	313	6,261
`bov`	BOV	5,465	303	303	6,071
`box`	BOX	5,039	279	279	5,597
`bqc`	BQC	16,596	921	921	18,438
`bqj`	BQJ	6,249	347	347	6,943
`bqp`	BQP	11,904	661	661	13,226
`bri`	BRI	5,863	325	325	6,513
`bsc`	BSC	7,166	398	398	7,962
`bsp`	BSP	8,117	450	450	9,017
`bsq`	BSQ	19,965	1,109	1,109	22,183
`bss`	BSS	5,610	311	311	6,232
`bst`	BST	3,785	210	210	4,205
`bsw`	BSW	2,009	111	111	2,231
`btg`	BTG	5,703	316	316	6,335
`btt`	BTT	5,499	305	305	6,109
`bud`	BUD	18,422	1,023	1,023	20,468
`bui`	BUI	6,163	342	342	6,847
`buj`	BUJ	774	42	42	858
`bum`	BUM	85,158	4,731	4,731	94,620
`bus`	BUS	11,456	636	636	12,728
`buu`	BUU	10,590	588	588	11,766
`buy`	BUY	867	48	48	963
`bvm`	BVM	5,538	307	307	6,152
`bwc`	BWC	885	49	49	983
`bwj`	BWJ	5,479	304	304	6,087
`bwo`	BWO	3,405	189	189	3,783
`bwr`	BWR	22,176	1,231	1,231	24,638
`bwt`	BWT	5,748	319	319	6,386
`bwu`	BWU	20,985	1,165	1,165	23,315
`bxk`	BXK	22,165	1,231	1,231	24,627
`byv`	BYV	25,382	1,410	1,410	28,202
`bza`	BZA	5,371	298	298	5,967
`bzw`	BZW	22,628	1,257	1,257	25,142
`cce`	CCE	23,256	1,291	1,291	25,838
`cen`	CEN	7,426	412	412	8,250
`cgg`	CGG	19,611	1,089	1,089	21,789
`cjk`	CJK	42,318	2,351	2,351	47,020
`ckl`	CKL	6,278	348	348	6,974
`cko`	CKO	4,950	274	274	5,498
`cky`	CKY	3,158	175	175	3,508
`cla`	CLA	5,197	288	288	5,773
`cll`	CLL	3,039	168	168	3,375
`cme`	CME	4,401	244	244	4,889
`cou`	COU	6,691	371	371	7,433
`crs`	CRS	126,872	7,048	7,048	140,968
`csk`	CSK	5,527	306	306	6,139
`cuv`	CUV	5,532	307	307	6,146
`cwa`	CWA	5,997	333	333	6,663
`cwe`	CWE	16,133	896	896	17,925
`cwt`	CWT	5,306	294	294	5,894
`daa`	DAA	4,836	268	268	5,372
`dae`	DAE	1,383	76	76	1,535
`daf`	DAF	37,909	2,106	2,106	42,121
`dag`	DAG	22,260	1,236	1,236	24,732
`dav`	DAV	22,798	1,266	1,266	25,330
`dbd`	DBD	5,311	294	294	5,899
`dde`	DDE	6,013	334	334	6,681
`dga`	DGA	42,933	2,385	2,385	47,703
`dgi`	DGI	20,394	1,133	1,133	22,660
`dhm`	DHM	5,133	285	285	5,703
`did`	DID	6,987	388	388	7,763
`dig`	DIG	21,789	1,210	1,210	24,209
`dik`	DIK	4,800	266	266	5,332
`dip`	DIP	20,219	1,123	1,123	22,465
`dks`	DKS	4,797	266	266	5,329
`dne`	DNE	11,509	639	639	12,787
`dnj`	DNJ	5,364	297	297	5,958
`dnn`	DNN	5,266	292	292	5,850
`dno`	DNO	3,558	197	197	3,952
`dow`	DOW	5,051	280	280	5,611
`dri`	DRI	4,273	237	237	4,747
`drs`	DRS	13,462	747	747	14,956
`dsh`	DSH	4,840	268	268	5,376
`dua`	DUA	50,532	2,807	2,807	56,146
`dug`	DUG	5,589	310	310	6,209
`dur`	DUR	4,640	257	257	5,154
`dwr`	DWR	11,651	647	647	12,945
`dyi`	DYI	21,526	1,195	1,195	23,916
`dyu`	Dyula / Jula	92,064	5,114	5,114	102,292
`ebr`	EBR	5,392	299	299	5,990
`ebu`	EBU	22,231	1,234	1,234	24,699
`ee`	Ewe	197,282	10,960	10,960	219,202
`efi`	Efik	174,619	9,701	9,701	194,021
`egm`	EGM	5,595	310	310	6,215
`ego`	EGO	5,809	322	322	6,453
`eka`	EKA	12,402	688	688	13,778
`ekm`	EKM	6,023	334	334	6,691
`eko`	EKO	7,852	436	436	8,724
`etu`	ETU	5,161	286	286	5,733
`etx`	ETX	6,024	334	334	6,692
`ewo`	Ewondo	13,319	739	739	14,797
`eza`	EZA	23,125	1,284	1,284	25,693
`fan`	Fang	22,716	1,262	1,262	25,240
`fat`	Fante / Akan	59,089	3,282	3,282	65,653
`ffm`	FFM	26,267	1,459	1,459	29,185
`fip`	FIP	4,972	276	276	5,524
`flr`	FLR	17,307	961	961	19,229
`fon`	Fon	70,652	3,925	3,925	78,502
`fub`	FUB	18,181	1,010	1,010	20,201
`fue`	FUE	20,829	1,157	1,157	23,143
`fuf`	Pular / Fula	9,985	554	554	11,093
`fuh`	FUH	5,602	311	311	6,224
`fuq`	FUQ	5,738	318	318	6,374
`fuv`	FUV	4,368	242	242	4,852
`gaa`	Ga	161,784	8,988	8,988	179,760
`gab`	GAB	5,942	330	330	6,602
`gax`	GAX	24,710	1,372	1,372	27,454
`gaz`	GAZ	23,977	1,332	1,332	26,641
`gba`	GBA	6,150	341	341	6,832
`gbo`	GBO	4,839	268	268	5,375
`gbr`	GBR	8,501	472	472	9,445
`gbv`	GBV	3,866	214	214	4,294
`gde`	GDE	5,192	288	288	5,768
`gdf`	GDF	1,327	73	73	1,473
`gej`	GEJ	18,284	1,015	1,015	20,314
`gel`	GEL	3,226	179	179	3,584
`gez`	GEZ	3,917	217	217	4,351
`gid`	GID	23,041	1,279	1,279	25,599
`giz`	GIZ	19,845	1,102	1,102	22,049
`gjn`	GJN	21,304	1,183	1,183	23,670
`gkn`	GKN	34,344	1,908	1,908	38,160
`gkp`	GKP	27,175	1,509	1,509	30,193
`glw`	GLW	2,043	113	113	2,269
`gmv`	GMV	6,624	368	368	7,360
`gna`	GNA	4,301	238	238	4,777
`gnd`	GND	4,979	276	276	5,531
`gng`	GNG	5,166	287	287	5,740
`goa`	GOA	19,919	1,106	1,106	22,131
`god`	GOD	4,950	275	275	5,500
`gof`	GOF	7,631	423	423	8,477
`gog`	GOG	24,473	1,359	1,359	27,191
`gol`	GOL	4,778	265	265	5,308
`gou`	GOU	5,802	322	322	6,446
`gqa`	GQA	2,451	136	136	2,723
`gqr`	GQR	5,095	283	283	5,661
`gsl`	GSL	1,638	91	91	1,820
`gso`	GSO	6,048	335	335	6,718
`gud`	GUD	5,850	325	325	6,500
`gur`	GUR	30,149	1,674	1,674	33,497
`guw`	GUW	112,403	6,244	6,244	124,891
`gux`	GUX	42,120	2,340	2,340	46,800
`guz`	GUZ	38,787	2,154	2,154	43,095
`gvl`	GVL	5,513	306	306	6,125
`gwn`	GWN	5,666	314	314	6,294
`gwr`	GWR	12,317	684	684	13,685
`gya`	GYA	18,950	1,052	1,052	21,054
`gyl`	GYL	3,522	195	195	3,912
`gyz`	GYZ	3,464	192	192	3,848
`ha`	Hausa	107,424	5,967	5,967	119,358
`hag`	HAG	1,034	57	57	1,148
`hav`	HAV	2,394	133	133	2,660
`hay`	HAY	19,327	1,073	1,073	21,473
`hbb`	HBB	5,628	312	312	6,252
`hbn`	HBN	5,542	307	307	6,156
`hdy`	HDY	13,428	745	745	14,918
`heh`	HEH	5,411	300	300	6,011
`her`	HER	19,040	1,057	1,057	21,154
`hwo`	HWO	5,779	321	321	6,421
`hz`	HZ	48,470	2,692	2,692	53,854
`ibb`	IBB	25,043	1,391	1,391	27,825
`iby`	IBY	5,670	315	315	6,300
`ich`	ICH	5,865	325	325	6,515
`idc`	IDC	2,826	157	157	3,140
`idu`	IDU	49,627	2,756	2,756	55,139
`ife`	IFE	4,266	237	237	4,740
`ig`	Igbo	194,294	10,794	10,794	215,882
`igb`	IGB	21,641	1,202	1,202	24,045
`ige`	IGE	51,161	2,842	2,842	56,845
`igl`	IGL	24,370	1,353	1,353	27,076
`ijn`	IJN	21,132	1,173	1,173	23,478
`ikk`	IKK	22,692	1,260	1,260	25,212
`ikw`	IKW	6,031	334	334	6,699
`ikz`	IKZ	5,641	313	313	6,267
`ilb`	ILB	5,839	324	324	6,487
`iqw`	IQW	22,857	1,269	1,269	25,395
`iri`	IRI	11,286	627	627	12,540
`irk`	IRK	24,692	1,371	1,371	27,434
`isn`	ISN	3,922	217	217	4,356
`iyx`	IYX	6,150	341	341	6,832
`izr`	IZR	5,739	318	318	6,375
`izz`	IZZ	22,970	1,276	1,276	25,522
`jab`	JAB	6,117	339	339	6,795
`jbn`	JBN	3,528	196	196	3,920
`jgo`	JGO	4,577	254	254	5,085
`jib`	JIB	5,527	307	307	6,141
`jid`	JID	3,985	221	221	4,427
`jit`	JIT	9,381	521	521	10,423
`jmc`	JMC	21,853	1,213	1,213	24,279
`jnj`	JNJ	2,645	146	146	2,937
`jow`	JOW	5,997	333	333	6,663
`jub`	JUB	721	40	40	801
`jum`	JUM	3,075	170	170	3,415
`kab`	KAB	38,176	2,120	2,120	42,416
`kaj`	KAJ	4,831	268	268	5,367
`kam`	KAM	107,325	5,962	5,962	119,249
`kbn`	KBN	6,197	344	344	6,885
`kbo`	KBO	7,406	411	411	8,228
`kbp`	KBP	64,943	3,607	3,607	72,157
`kbr`	KBR	17,534	974	974	19,482
`kby`	KBY	4,150	230	230	4,610
`kcc`	KCC	607	33	33	673
`kcg`	KCG	5,374	298	298	5,970
`kck`	KCK	40,430	2,246	2,246	44,922
`kcq`	KCQ	5,205	289	289	5,783
`kcu`	KCU	4,342	241	241	4,824
`kdc`	KDC	14,098	783	783	15,664
`kde`	KDE	16,878	937	937	18,752
`kdh`	KDH	6,614	367	367	7,348
`kdi`	KDI	22,919	1,273	1,273	25,465
`kdj`	KDJ	24,089	1,338	1,338	26,765
`kdl`	KDL	5,370	298	298	5,966
`kdn`	KDN	6,678	371	371	7,420
`kea`	KEA	55,116	3,061	3,061	61,238
`ked`	KED	4,913	272	272	5,457
`ken`	KEN	4,725	262	262	5,249
`keo`	KEO	5,826	323	323	6,472
`ker`	KER	5,072	281	281	5,634
`khy`	KHY	6,147	341	341	6,829
`ki`	KI	77,491	4,305	4,305	86,101
`kia`	KIA	24,340	1,352	1,352	27,044
`kik`	KIK	63,967	3,553	3,553	71,073
`kin`	Kinyarwanda	182,220	10,123	10,123	202,466
`kiv`	KIV	1,983	110	110	2,203
`kki`	KKI	5,038	279	279	5,596
`kkj`	KKJ	5,869	326	326	6,521
`kln`	KLN	36,862	2,047	2,047	40,956
`klu`	KLU	18,836	1,046	1,046	20,928
`kma`	KMA	4,844	269	269	5,382
`kmb`	KMB	84,038	4,668	4,668	93,374
`kmw`	KMW	5,280	293	293	5,866
`knf`	KNF	8,069	448	448	8,965
`knk`	KNK	18,427	1,023	1,023	20,473
`kno`	KNO	4,408	244	244	4,896
`kon`	KON	102,696	5,705	5,705	114,106
`kpz`	KPZ	16,831	935	935	18,701
`kqn`	KQN	6,229	345	345	6,919
`kqo`	KQO	5,358	297	297	5,952
`kqp`	KQP	5,580	310	310	6,200
`kqs`	KQS	21,076	1,170	1,170	23,416
`kqy`	KQY	3,669	203	203	4,075
`kri`	KRI	81,612	4,534	4,534	90,680
`krs`	KRS	7,183	398	398	7,979
`krw`	KRW	5,077	282	282	5,641
`krx`	KRX	5,254	291	291	5,836
`ksb`	KSB	5,499	305	305	6,109
`ksf`	KSF	4,986	277	277	5,540
`ksp`	KSP	7,872	437	437	8,746
`kss`	KSS	13,907	772	772	15,451
`ktb`	KTB	14,846	824	824	16,494
`ktj`	KTJ	4,570	253	253	5,076
`ktu`	KTU	32,923	1,829	1,829	36,581
`kua`	KUA	86,023	4,779	4,779	95,581
`kub`	KUB	8,440	468	468	9,376
`kuj`	KUJ	23,202	1,288	1,288	25,778
`kus`	KUS	21,504	1,194	1,194	23,892
`kvj`	KVJ	18,829	1,046	1,046	20,921
`kwm`	KWM	4,969	276	276	5,521
`kwn`	KWN	16,554	919	919	18,392
`kwy`	KWY	83,685	4,649	4,649	92,983
`kxc`	KXC	19,691	1,093	1,093	21,877
`kya`	KYA	5,825	323	323	6,471
`kyf`	KYF	5,200	288	288	5,776
`kyq`	KYQ	5,207	289	289	5,785
`kzr`	KZR	5,063	281	281	5,625
`lag`	LAG	5,799	322	322	6,443
`lai`	LAI	20,892	1,160	1,160	23,212
`laj`	LAJ	26,084	1,449	1,449	28,982
`lam`	LAM	17,044	946	946	18,936
`lap`	LAP	7,167	398	398	7,963
`ldb`	LDB	6,142	341	341	6,824
`ldi`	LDI	31,574	1,754	1,754	35,082
`ldk`	LDK	5,080	282	282	5,644
`leb`	LEB	5,007	278	278	5,563
`led`	LED	24,114	1,339	1,339	26,792
`lee`	LEE	4,836	268	268	5,372
`lef`	LEF	19,472	1,081	1,081	21,634
`leh`	LEH	19,459	1,081	1,081	21,621
`lem`	LEM	5,915	328	328	6,571
`lgg`	LGG	28,683	1,593	1,593	31,869
`lgm`	LGM	5,940	330	330	6,600
`lia`	LIA	6,192	343	343	6,878
`lik`	LIK	5,575	309	309	6,193
`lin`	Lingala	124,932	6,940	6,940	138,812
`lip`	LIP	5,298	294	294	5,886
`liy`	LIY	5,599	310	310	6,219
`lla`	LLA	4,139	229	229	4,597
`llb`	LLB	14,492	805	805	16,102
`lmd`	LMD	5,232	290	290	5,812
`lme`	LME	6,033	335	335	6,703
`lmp`	LMP	5,969	331	331	6,631
`lnl`	LNL	5,657	314	314	6,285
`lns`	LNS	20,772	1,154	1,154	23,080
`lob`	LOB	4,699	260	260	5,219
`log`	LOG	5,081	282	282	5,645
`lol`	LOL	27,007	1,500	1,500	30,007
`lom`	LOM	5,163	286	286	5,735
`lon`	LON	21,101	1,172	1,172	23,445
`loq`	LOQ	5,610	311	311	6,232
`lot`	LOT	11,153	619	619	12,391
`loz`	LOZ	164,339	9,129	9,129	182,597
`lsm`	LSM	21,188	1,177	1,177	23,542
`lth`	LTH	6,551	363	363	7,277
`lto`	LTO	22,662	1,259	1,259	25,180
`lua`	LUA	162,816	9,045	9,045	180,906
`lub`	LUB	113,104	6,283	6,283	125,670
`luc`	LUC	13,275	737	737	14,749
`lue`	LUE	23,654	1,314	1,314	26,282
`lug`	LUG	176,145	9,785	9,785	195,715
`luj`	LUJ	6,174	343	343	6,860
`lun`	LUN	36,127	2,007	2,007	40,141
`luo`	LUO	125,212	6,956	6,956	139,124
`lwg`	LWG	37,286	2,071	2,071	41,428
`lwo`	LWO	4,097	227	227	4,551
`lyn`	LYN	5,885	326	326	6,537
`maf`	MAF	25,747	1,430	1,430	28,607
`mas`	MAS	41,286	2,293	2,293	45,872
`maw`	MAW	5,269	292	292	5,853
`mbu`	MBU	19,620	1,090	1,090	21,800
`mck`	MCK	21,997	1,222	1,222	24,441
`mcn`	MCN	21,075	1,170	1,170	23,415
`mcp`	MCP	6,273	348	348	6,969
`mcu`	MCU	4,783	265	265	5,313
`mda`	MDA	5,023	278	278	5,579
`mdm`	MDM	5,110	283	283	5,676
`mdy`	MDY	20,988	1,165	1,165	23,318
`men`	MEN	5,060	281	281	5,622
`meq`	MEQ	5,046	280	280	5,606
`mer`	MER	37,740	2,096	2,096	41,932
`mev`	MEV	4,770	265	265	5,300
`mey`	MEY	1,169	64	64	1,297
`mfe`	MFE	96,049	5,336	5,336	106,721
`mfg`	MFG	6,989	388	388	7,765
`mfh`	MFH	5,508	305	305	6,118
`mfi`	MFI	6,210	345	345	6,900
`mfk`	MFK	5,650	313	313	6,276
`mfm`	MFM	4,135	229	229	4,593
`mfn`	MFN	8,639	479	479	9,597
`mfo`	MFO	6,539	363	363	7,265
`mfq`	MFQ	7,585	421	421	8,427
`mfv`	MFV	6,277	348	348	6,973
`mfx`	MFX	3,366	187	187	3,740
`mgc`	MGC	2,808	156	156	3,120
`mge`	MGE	5,383	298	298	5,979
`mgg`	MGG	5,448	302	302	6,052
`mgh`	MGH	34,551	1,919	1,919	38,389
`mgj`	MGJ	2,928	162	162	3,252
`mgo`	MGO	8,538	474	474	9,486
`mgq`	MGQ	5,279	293	293	5,865
`mgr`	MGR	80,980	4,498	4,498	89,976
`mgs`	MGS	6,164	342	342	6,848
`mgv`	MGV	11,880	660	660	13,200
`mgw`	MGW	5,761	320	320	6,401
`mgy`	MGY	5,671	314	314	6,299
`mhd`	MHD	1,975	109	109	2,193
`mho`	MHO	6,022	334	334	6,690
`mhw`	MHW	4,660	258	258	5,176
`mif`	MIF	5,782	321	321	6,424
`mkl`	MKL	18,037	1,002	1,002	20,041
`mkw`	MKW	22,311	1,239	1,239	24,789
`mlg`	MLG	235,396	13,077	13,077	261,550
`mlw`	MLW	5,868	326	326	6,520
`mmu`	MMU	5,398	299	299	5,996
`mmy`	MMY	5,328	296	296	5,920
`mnf`	MNF	4,042	224	224	4,490
`mnh`	MNH	3,889	215	215	4,319
`mnk`	MNK	23,419	1,300	1,300	26,019
`moa`	MOA	5,779	320	320	6,419
`mos`	MOS	123,787	6,877	6,877	137,541
`moy`	MOY	4,930	273	273	5,476
`moz`	MOZ	5,901	327	327	6,555
`mpa`	MPA	5,890	327	327	6,544
`mpe`	MPE	4,757	264	264	5,285
`mpg`	MPG	19,466	1,081	1,081	21,628
`mqb`	MQB	5,251	291	291	5,833
`msc`	MSC	7,689	427	427	8,543
`mse`	MSE	20,011	1,111	1,111	22,233
`mua`	MUA	18,829	1,046	1,046	20,921
`mug`	MUG	19,421	1,078	1,078	21,577
`muh`	MUH	5,492	305	305	6,102
`mur`	MUR	6,396	355	355	7,106
`muy`	MUY	5,286	293	293	5,872
`mwe`	MWE	5,584	310	310	6,204
`mwm`	MWM	23,763	1,320	1,320	26,403
`mwn`	MWN	17,566	975	975	19,516
`mxx`	MXX	3,450	191	191	3,832
`myb`	MYB	35,280	1,959	1,959	39,198
`mye`	MYE	19,817	1,100	1,100	22,017
`myk`	MYK	6,039	335	335	6,709
`myx`	MYX	27,703	1,539	1,539	30,781
`mzj`	MZJ	5,694	316	316	6,326
`mzm`	MZM	4,960	275	275	5,510
`mzw`	MZW	19,576	1,087	1,087	21,750
`naq`	NAQ	53,490	2,971	2,971	59,432
`nat`	NAT	5,014	278	278	5,570
`naw`	NAW	6,073	337	337	6,747
`nbl`	NBL	15,766	875	875	17,516
`ncr`	NCR	5,819	323	323	6,465
`ncu`	NCU	4,662	259	259	5,180
`nd`	ND	60,723	3,373	3,373	67,469
`ndc`	NDC	71,110	3,950	3,950	79,010
`nde`	NDE	29,918	1,662	1,662	33,242
`ndh`	NDH	5,304	294	294	5,892
`ndi`	NDI	5,059	281	281	5,621
`ndj`	NDJ	5,895	327	327	6,549
`ndo`	NDO	77,247	4,291	4,291	85,829
`ndp`	NDP	5,693	316	316	6,325
`ndv`	NDV	6,527	362	362	7,251
`ndz`	NDZ	5,409	300	300	6,009
`ney`	NEY	2,268	126	126	2,520
`nga`	NGA	16,385	910	910	18,205
`ngb`	NGB	26,299	1,460	1,460	29,219
`ngc`	NGC	12,631	701	701	14,033
`nge`	NGE	5,943	330	330	6,603
`ngj`	NGJ	5,952	330	330	6,612
`ngl`	NGL	53,630	2,979	2,979	59,588
`ngn`	NGN	6,042	335	335	6,712
`ngp`	NGP	19,045	1,058	1,058	21,161
`ngz`	NGZ	6,026	334	334	6,694
`nhr`	NHR	5,474	304	304	6,082
`nhu`	NHU	5,952	330	330	6,612
`nie`	NIE	4,984	276	276	5,536
`nih`	NIH	5,333	296	296	5,925
`nim`	NIM	4,706	261	261	5,228
`nin`	NIN	5,665	314	314	6,293
`niy`	NIY	5,889	327	327	6,543
`njd`	NJD	4,140	229	229	4,598
`nka`	NKA	27,282	1,515	1,515	30,312
`nkn`	NKN	10,210	567	567	11,344
`nko`	NKO	6,720	373	373	7,466
`nla`	NLA	5,047	280	280	5,607
`nle`	NLE	7,009	389	389	7,787
`nmq`	NMQ	4,889	271	271	5,431
`nnb`	NNB	22,562	1,253	1,253	25,068
`nnh`	NNH	29,253	1,625	1,625	32,503
`nnq`	NNQ	5,523	306	306	6,135
`nnw`	NNW	5,171	287	287	5,745
`nr`	NR	57,604	3,200	3,200	64,004
`nse`	NSE	9,504	527	527	10,558
`nso`	NSO	81,036	4,502	4,502	90,040
`nsx`	NSX	4,321	239	239	4,799
`ntk`	NTK	5,627	312	312	6,251
`ntr`	NTR	5,544	307	307	6,158
`nuj`	NUJ	11,167	620	620	12,407
`nus`	NUS	19,331	1,073	1,073	21,477
`nwb`	NWB	4,194	233	233	4,660
`nww`	NWW	5,784	321	321	6,426
`nxd`	NXD	5,373	298	298	5,969
`nya`	Chichewa	127,440	7,080	7,080	141,600
`nyb`	NYB	1,234	68	68	1,370
`nyd`	NYD	18,788	1,043	1,043	20,874
`nyf`	NYF	5,689	316	316	6,321
`nyj`	NYJ	2,375	131	131	2,637
`nym`	NYM	8,478	471	471	9,420
`nyn`	NYN	97,244	5,402	5,402	108,048
`nyo`	NYO	36,609	2,033	2,033	40,675
`nyu`	NYU	8,316	462	462	9,240
`nyy`	NYY	32,000	1,777	1,777	35,554
`nza`	NZA	4,989	277	277	5,543
`nzi`	NZI	99,578	5,532	5,532	110,642
`odu`	ODU	5,172	287	287	5,746
`okr`	OKR	20,281	1,126	1,126	22,533
`oku`	OKU	5,850	324	324	6,498
`old`	OLD	20,343	1,130	1,130	22,603
`omi`	OMI	7,020	389	389	7,798
`orm`	ORM	90,864	5,047	5,047	100,958
`otr`	OTR	5,421	301	301	6,023
`oyd`	OYD	3,900	216	216	4,332
`ozm`	OZM	6,522	362	362	7,246
`pcm`	PCM	23,814	1,322	1,322	26,458
`pfe`	PFE	5,376	298	298	5,972
`pkb`	PKB	5,592	310	310	6,212
`pko`	PKO	20,324	1,129	1,129	22,582
`plt`	PLT	89,669	4,981	4,981	99,631
`pny`	PNY	5,576	309	309	6,194
`pnz`	PNZ	5,741	318	318	6,377
`por`	POR	262,339	14,574	14,574	291,487
`pov`	POV	24,152	1,341	1,341	26,834
`poy`	POY	19,956	1,108	1,108	22,172
`puu`	PUU	20,002	1,111	1,111	22,224
`rag`	RAG	37,618	2,089	2,089	41,796
`rcf`	RCF	21,512	1,195	1,195	23,902
`rel`	REL	5,388	299	299	5,986
`rif`	RIF	5,103	283	283	5,669
`rim`	RIM	5,994	332	332	6,658
`rng`	RNG	45,615	2,534	2,534	50,683
`rub`	RUB	8,569	475	475	9,519
`ruc`	RUC	4,702	261	261	5,224
`ruf`	RUF	17,935	996	996	19,927
`run`	Kirundi	129,888	7,216	7,216	144,320
`rwk`	RWK	5,660	314	314	6,288
`rwm`	RWM	1,623	90	90	1,803
`saa`	SAA	1,167	64	64	1,295
`sag`	SAG	99,420	5,523	5,523	110,466
`saq`	SAQ	6,329	351	351	7,031
`sav`	SAV	7,094	394	394	7,882
`sba`	SBA	21,348	1,185	1,185	23,718
`sbd`	SBD	19,062	1,058	1,058	21,178
`sbk`	SBK	5,116	284	284	5,684
`sbm`	SBM	4,554	252	252	5,058
`sbp`	SBP	6,224	345	345	6,914
`sby`	SBY	5,454	302	302	6,058
`sef`	SEF	21,489	1,193	1,193	23,875
`seh`	SEH	64,787	3,599	3,599	71,985
`ses`	SES	18,090	1,004	1,004	20,098
`sev`	SEV	11,698	649	649	12,996
`sfw`	SFW	28,722	1,595	1,595	31,912
`sgw`	SGW	11,134	618	618	12,370
`shi`	SHI	10,652	591	591	11,834
`shk`	SHK	19,981	1,110	1,110	22,201
`shq`	SHQ	1,497	83	83	1,663
`shr`	SHR	17,245	958	958	19,161
`shu`	SHU	14,806	822	822	16,450
`sid`	SID	66,119	3,673	3,673	73,465
`sig`	SIG	4,986	276	276	5,538
`sil`	SIL	20,516	1,139	1,139	22,794
`sn`	Shona	199,226	11,068	11,068	221,362
`snf`	SNF	4,711	261	261	5,233
`snm`	SNM	1,611	89	89	1,789
`snw`	SNW	19,169	1,064	1,064	21,297
`som`	SOM	22,349	1,241	1,241	24,831
`sop`	SOP	5,285	293	293	5,871
`sor`	SOR	3,421	189	189	3,799
`sot`	SOT	104,631	5,812	5,812	116,255
`soy`	SOY	5,141	285	285	5,711
`spa`	SPA	323,858	17,992	17,992	359,842
`spp`	SPP	6,930	384	384	7,698
`sqa`	SQA	1,712	95	95	1,902
`srr`	SRR	22,789	1,265	1,265	25,319
`ssc`	SSC	5,851	325	325	6,501
`ssw`	SSW	82,465	4,581	4,581	91,627
`ssy`	SSY	5,338	296	296	5,930
`suk`	SUK	29,646	1,646	1,646	32,938
`suq`	SUQ	3,420	190	190	3,800
`sus`	SUS	19,596	1,088	1,088	21,772
`sw`	Swahili	322,541	17,918	17,918	358,377
`swb`	SWB	5,801	322	322	6,445
`swc`	SWC	25,493	1,416	1,416	28,325
`swk`	SWK	22,526	1,251	1,251	25,028
`sxb`	SXB	5,741	318	318	6,377
`szv`	SZV	5,496	305	305	6,106
`tan`	TAN	2,663	147	147	2,957
`tap`	TAP	4,009	222	222	4,453
`taq`	TAQ	4,447	246	246	4,939
`tbt`	TBT	5,292	294	294	5,880
`tcc`	TCC	25,074	1,393	1,393	27,860
`tcd`	TCD	5,488	304	304	6,096
`tdx`	TDX	44,150	2,452	2,452	49,054
`ted`	TED	4,438	246	246	4,930
`tem`	TEM	5,455	303	303	6,061
`teo`	TEO	49,881	2,771	2,771	55,423
`tex`	TEX	6,576	365	365	7,306
`tga`	TGA	5,982	332	332	6,646
`tgw`	TGW	20,420	1,134	1,134	22,688
`thk`	THK	5,725	317	317	6,359
`thv`	THV	2,091	116	116	2,323
`tir`	TIR	159,744	8,874	8,874	177,492
`tiv`	TIV	130,725	7,262	7,262	145,249
`tke`	TKE	7,793	432	432	8,657
`tlj`	TLJ	14,051	780	780	15,611
`tll`	TLL	22,896	1,271	1,271	25,438
`tng`	TNG	5,603	311	311	6,225
`tnr`	TNR	4,606	255	255	5,116
`tod`	TOD	18,910	1,050	1,050	21,010
`tog`	TOG	91,236	5,068	5,068	101,372
`toh`	TOH	22,528	1,251	1,251	25,030
`toi`	TOI	152,100	8,450	8,450	169,000
`tpm`	TPM	4,036	224	224	4,484
`tsa`	TSA	6,242	346	346	6,934
`tsc`	TSC	24,070	1,337	1,337	26,744
`tsn`	TSN	235,958	13,108	13,108	262,174
`tso`	TSO	190,999	10,611	10,611	212,221
`tsv`	TSV	5,234	290	290	5,814
`tsw`	TSW	21,916	1,217	1,217	24,350
`ttj`	TTJ	50,431	2,801	2,801	56,033
`ttq`	TTQ	4,928	273	273	5,474
`ttr`	TTR	5,657	314	314	6,285
`tui`	TUI	35,680	1,982	1,982	39,644
`tul`	TUL	4,824	267	267	5,358
`tum`	TUM	130,431	7,246	7,246	144,923
`tuv`	TUV	23,016	1,278	1,278	25,572
`tvs`	TVS	8,195	455	455	9,105
`tvu`	TVU	5,995	332	332	6,659
`tw`	Twi / Akan	580,129	32,229	32,229	644,587
`tyx`	TYX	5,898	327	327	6,552
`udu`	UDU	8,839	491	491	9,821
`umb`	UMB	92,070	5,115	5,115	102,300
`urh`	URH	24,568	1,364	1,364	27,296
`uth`	UTH	4,175	231	231	4,637
`utr`	UTR	6,059	336	336	6,731
`vag`	VAG	20,699	1,149	1,149	22,997
`vai`	VAI	4,060	225	225	4,510
`ven`	VEN	110,937	6,163	6,163	123,263
`ver`	VER	3,450	191	191	3,832
`vid`	VID	16,963	942	942	18,847
`vif`	VIF	6,202	344	344	6,890
`vmk`	VMK	46,525	2,584	2,584	51,693
`vmw`	VMW	55,579	3,087	3,087	61,753
`vun`	VUN	5,678	315	315	6,308
`vut`	VUT	5,729	318	318	6,365
`wal`	WAL	81,948	4,552	4,552	91,052
`wbi`	WBI	5,287	293	293	5,873
`weh`	WEH	5,383	299	299	5,981
`wib`	WIB	4,101	227	227	4,555
`wja`	WJA	1,262	70	70	1,402
`wmw`	WMW	11,284	626	626	12,536
`wob`	WOB	5,191	288	288	5,767
`wof`	WOF	1,596	88	88	1,772
`wol`	WOL	39,237	2,179	2,179	43,595
`xan`	XAN	3,373	187	187	3,747
`xed`	XED	16,274	904	904	18,082
`xho`	isiXhosa	170,587	9,476	9,476	189,539
`xmc`	XMC	12,165	675	675	13,515
`xmd`	XMD	5,831	323	323	6,477
`xmg`	XMG	5,748	319	319	6,386
`xmv`	XMV	47,852	2,658	2,658	53,168
`xnj`	XNJ	7,862	436	436	8,734
`xnz`	XNZ	2,241	124	124	2,489
`xog`	XOG	19,974	1,109	1,109	22,192
`xon`	XON	18,615	1,034	1,034	20,683
`xpe`	XPE	32,616	1,812	1,812	36,240
`xrb`	XRB	4,805	266	266	5,337
`xsm`	XSM	22,788	1,265	1,265	25,318
`xsn`	XSN	1,640	91	91	1,822
`xtc`	XTC	6,109	339	339	6,787
`xuo`	XUO	5,429	301	301	6,031
`yaf`	YAF	5,320	295	295	5,910
`yal`	YAL	19,692	1,094	1,094	21,880
`yam`	YAM	5,042	280	280	5,602
`yao`	YAO	90,360	5,020	5,020	100,400
`yas`	YAS	5,562	308	308	6,178
`yat`	YAT	5,391	299	299	5,989
`yav`	YAV	4,874	270	270	5,414
`yba`	YBA	5,569	309	309	6,187
`ybb`	YBB	9,430	523	523	10,476
`yns`	YNS	6,034	335	335	6,704
`yo`	Yoruba	212,259	11,792	11,792	235,843
`yom`	YOM	11,181	621	621	12,423
`yre`	YRE	5,170	287	287	5,744
`zaj`	ZAJ	11,638	646	646	12,930
`zak`	ZAK	5,612	311	311	6,234
`zga`	ZGA	5,020	278	278	5,576
`zin`	ZIN	5,619	312	312	6,243
`ziw`	ZIW	5,655	314	314	6,283
`zne`	ZNE	81,798	4,544	4,544	90,886
`zul`	isiZulu	150,150	8,341	8,341	166,832
Total		17,530,453	973,542	973,542	19,477,537

How to Load Subsets

You can load any specific African language subset directly with Hugging Face datasets:

from datasets import load_dataset

# 1. Load Twi (Akan) subset
dataset_tw = load_dataset("AfriSpeech/gloss-mt-train-en", "tw")
print(dataset_tw["train"][0])

# 2. Load Swahili subset
dataset_sw = load_dataset("AfriSpeech/gloss-mt-train-en", "sw")
print(dataset_sw["train"][0])

# 3. Load Amharic subset (if available)
dataset_amh = load_dataset("AfriSpeech/gloss-mt-train-en", "amh")
print(dataset_amh["train"][0])

Data Fields

input: The source sentence with in-place target glossary annotations (e.g. The <kannifoɔ> made a covenant...).
tgt: The clean target sentence (e.g. Na kannifoɔ no ne Nyankopɔn yɛɛ apam...).
src_raw: Original clean source sentence before glossary injection.
tgt_ref: Original target reference sentence.
inserted_terms: List of glossary terms injected into the source text.
src_lang: Source language code.
tgt_lang: Target language code.

Citation

@article{owusu2026afrigloss,
  title={Afri-Gloss-MT: In-Place Glossary Constraints for African Machine Translation},
  author={Owusu, Mich-Seth},
  journal={arXiv preprint},
  year={2026}
}