Fon - UNICEF WCARO NLP

🌍 General Information

ISO 639-3fon

ISO 639-1—

French Namefon

Glottocodefonn1241

Language FamilyNiger–Congo

Writing SystemLatin Gbékoun

L1 Speakers2.3 million

L2 Speakers—

Population2300000 (10⁶+)

Endangermentsafe

CountriesBenin, Togo

Also Known AsDahomeen, Djedji, Fo, Fogbe, Fon-Gbe, Fongbe, Fonnu, Fɔngbè

WikipediaFon on Wikipedia

🗺️ Language Use

Data from CLEAR Global's Language Use Data Platform.

👥 Actors (4)

Farmerline / Darli AI (Startup)

Ghanaian agritech company building technology solutions for smallholder farmers. Core platform Mergdata is a cloud-based farm management system used b...

Institut de Mathématiques et de Sciences Physiques (IMSP) (University)

Research institute affiliated with University of Abomey-Calavi, offering programs in mathematics, physics, data science, and information technologies....

Masakhane (Research Community And Foundation)

Masakhane ("We build together" in isiZulu) is a grassroots NLP organization dedicated to strengthening natural language processing research in African...

UBC Deep Learning & NLP Lab (University)

Research lab at the University of British Columbia led by Muhammad Abdul-Mageed, focused on deep representation learning and natural language socio-pr...

🤖 NLP & Tech Resources

✓ AfroLID

✓ Google Translate: 2024

✓ NLLB-200

✓ MADLAD-400 Docs: 1100 MAD-Docs

✓ MADLAD-400 Sentences: 67300 MAD-Sents

✓ Fineweb2: 💬

✓ CMU Wilderness

✓ MAFAND

✓ MMS-ASR

✓ MMS-TTS

✓ MMS-LID

✓ Omnilingual ASR

✓ SimbaBench

Benchmarks

ASR (Speech-to-Text)

SimbaBench (ALFFA)

Model	WER	CER	Source
facebook/mms-1b-all	86.21	31.62	reported
openai/whisper-large-v3	344.41	312.17	reported
openai/whisper-large-v3-turbo	135.9	124.25	reported
facebook/seamless-m4t-v2-large	132.78	115.56	reported
UBC-NLP/Simba-S	42.12	14.29	reported
UBC-NLP/Simba-M	42.49	11.77	reported
UBC-NLP/omniASR_LLM_7B_v2	86.71	35.01	reported
UBC-NLP/omniASR_LLM_1B_v2	85.11	33.49	reported
UBC-NLP/omniASR_LLM_300M_v2	82.84	31.67	reported
UBC-NLP/omniASR_CTC_7B_v2	83.29	32.92	reported
UBC-NLP/omniASR_CTC_3B_v2	85.37	33.51	reported
UBC-NLP/omniASR_CTC_1B_v2	90.6	37.36	reported

MT (Machine Translation)

FLORES+ 10-sample

Model	direction	SpBLEU	ChrF	Source
anthropic/claude-opus-4.5	fon→eng	12.0	34.0	reported
anthropic/claude-opus-4.5	eng→fon	7.7	26.4	reported
anthropic/claude-opus-4.6	fon→eng	11.6	34.7	reported
anthropic/claude-opus-4.6	eng→fon	4.2	23.3	reported
google/gemini-2.5-pro	eng→fon	6.9	26.0	reported
google/gemini-2.5-pro	fon→eng	12.2	32.3	reported
google/gemini-3-pro-preview	fon→eng	18.7	41.5	reported
google/gemini-3-pro-preview	eng→fon	7.5	27.6	reported

FLORES-200

Model	direction	BLEU	chrF	Source
google/madlad400-3b-mt	fon→eng	3.9	25.1	reported
google/madlad400-3b-mt	eng→fon	1.6	15.6	reported
google/madlad400-7.2b-mt	fon→eng	5.6	26.3	reported
google/madlad400-7.2b-mt	eng→fon	1.9	16	reported
google/madlad400-10.7b-mt	fon→eng	6.4	27.8	reported
google/madlad400-10.7b-mt	eng→fon	1.6	15.9	reported
google/madlad400-8b-lm	fon→eng	1.8	18.3	reported
google/madlad400-8b-lm	eng→fon	0	1.6	reported

LLM (Large Language Models)

FLORES-200

Model	Log-perplexity	Source
goldfish-models/fon_latn_full	190.75	reported
MaLA-LM/mala-500-10b-v2	190.66	reported

SIB-200 10-sample

Model	Classification Acc.	Source
anthropic/claude-3.7-sonnet	100.0	reported
anthropic/claude-opus-4.5	100.0	reported
anthropic/claude-sonnet-4.6	100.0	reported

SAHARA

Model	Lang. Avg.	spBLEU fra→fon	NER F1	Source
gemini-3-pro-preview	30.3	8.74	51.85	reported
gemini-2.5-pro	29.66	7.27	52.05	reported
gpt-5-2025-08-07	21.92	2.81	41.03	reported
gpt-5.1-2025-11-13	16.26	3.19	29.33	reported
gemini-2.5-flash	6.71	2.15	11.27	reported
claude-sonnet-4-5@20250929	13.21	1.81	24.62	reported
claude-4-sonnet-20250514	12.54	0.46	24.62	reported
gpt-4.1	6.26	1.09	11.43	reported
CohereForAI/c4ai-command-a-03-2025	7.29	1.25	13.33	reported
meta-llama/Llama-4-Scout-17B-16E-Instruct	4.88	1.07	8.7	reported
meta-llama/Llama-3.3-70B-Instruct	4.05	0.7	7.41	reported
google/gemma-3-27b-it	3.68	1.39	5.97	reported
google/gemma-2-27b-it	5.41	1.05	9.76	reported
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	7.64	0.64	14.63	reported
meta-llama/Llama-3.1-70B-Instruct	3.44	0.64	6.25	reported
Qwen/Qwen3-30B-Instruct-2507	0.1	0.21	0.0	reported
deepseek-ai/DeepSeek-R1-Distill-Llama-70B	3.18	0.48	5.88	reported
Tower-Babel/Babel-9B-Chat	6.04	0.31	11.76	reported
Tower-Babel/Babel-83B-Chat	5.11	0.22	10.0	reported
Qwen/Qwen3-4B-Instruct-2507	0.15	0.3	0.0	reported
google/gemma-2-9b-it	2.23	0.29	4.17	reported
google/gemma-3-12b-it	2.92	0.95	4.88	reported
meta-llama/Llama-3.1-8B-Instruct	0.09	0.17	0.0	reported
CohereForAI/aya-23-35B	0.1	0.21	0.0	reported
CohereForAI/c4ai-command-r-plus-08-2024	0.07	0.14	0.0	reported
CohereForAI/c4ai-command-r7b-12-2024	0.2	0.41	0.0	reported
microsoft/Phi-4-mini-instruct	0.3	0.6	0.0	reported
meta-llama/Llama-3.2-1B-Instruct	0.23	0.47	0.0	reported
microsoft/Phi-3.5-mini-instruct	0.08	0.16	0.0	reported
google/gemma-2-2b-it	0.1	0.19	0.0	reported
CohereForAI/aya-23-8B	0.1	0.2	0.0	reported
google/gemma-3-4b-it	0.45	0.9	0.0	reported
meta-llama/Llama-3.2-3B-Instruct	0.13	0.26	0.0	reported

📝 Notes

Observations

Survey Focus 1 for Benin (2026 UNICEF WCAR Local Languages Survey)

ASR Models (18)

Model	Size	Downloads	Likes
speechbrain/asr-wav2vec2-dvoice-fongbe	—	972	3
espnet/xeus	—	60	146
asrDIL/mms-fongbe-finetuned-v4	1.0B	21	0
johntsi/ZeroSwot-Medium_asr-cv_en-to-200	0.4B	17	0
johntsi/ZeroSwot-Large_asr-cv_en-to-200	0.4B	16	1
chrisjay/fonxlsr	—	15	7
UBC-NLP/Simba-SLID-49	94.6M	15	0
johntsi/ZeroSwot-Medium_asr-mustc_en-to-200	0.4B	14	0
aioxlabs/dvoice-fongbe	—	12	1
johntsi/ZeroSwot-Large_asr-mustc_en-to-200	0.4B	12	0
asrDIL/fongbe-asr-v1.1.0	—	8	1
whettenr/asr-fon-french-streaming-conformer-without-diacritics	—	5	0
whettenr/asr-fon-streaming-conformer-without-diacritics	—	4	0
whettenr/asr-fon-with-diacritics-bpe-256	—	3	0
Professor/mms-300m-fongbe	0.3B	2	0
Paulalex/WhisperSmallFontranscript	0.2B	1	0
whettenr/asr-fon-with-diacritics	—	1	1
whettenr/asr-fon-without-diacritics	—	0	0

View on HuggingFace: fon: 18

TTS Models (1)

Model	Size	Downloads	Likes
Umbaji001/eyaa-tom-mms-tts-fon	36.3M	5	0

View on HuggingFace: fon: 1

MT Models (30)

Model	Size	Downloads	Likes
facebook/nllb-200-distilled-600M	—	1.1M	880
google/madlad400-3b-mt	3B	329.0K	192
facebook/nllb-200-distilled-1.3B	—	147.0K	181
facebook/nllb-200-3.3B	—	81.6K	443
jbochi/madlad400-3b-mt	3B	76.4K	154
facebook/nllb-200-1.3B	—	31.4K	71
Xenova/nllb-200-distilled-600M	—	4.4K	50
JustFrederik/nllb-200-distilled-600M-ct2-int8	—	925	1
facebook/nllb-moe-54b	—	557	123
JustFrederik/nllb-200-1.3B-ct2-int8	—	200	0
JustFrederik/nllb-200-distilled-600M-ct2-float16	—	157	0
winstxnhdw/nllb-200-distilled-1.3B-ct2-int8	—	141	6
JustFrederik/nllb-200-distilled-1.3B-ct2-int8	—	116	2
JustFrederik/nllb-200-distilled-600M-ct2	—	65	2
JustFrederik/nllb-200-3.3B-ct2-float16	—	52	3
michaelfeil/ct2fast-nllb-200-distilled-1.3B	—	47	1
michaelfeil/ct2fast-nllb-200-3.3B	—	43	14
ychenNLP/nllb-200-3.3B-easyproject	—	36	8
JustFrederik/nllb-200-distilled-1.3B-ct2-float16	—	30	1
JustFrederik/nllb-200-distilled-1.3B-ct2	—	29	0

View on HuggingFace: fon: 30

LLM Models (30)

Model	Size	Downloads	Likes
bigscience/bloomz-560m	0.6B	1.0M	137
bigscience/bloom-560m	0.6B	164.0K	371
bigscience/bloom-1b7	2B	49.4K	122
bigscience/bloom-3b	3B	14.1K	94
bigscience/bloomz-3b	3B	13.3K	81
bigscience/bloom-7b1	7B	10.4K	202
bigscience/bloom-1b1	1B	7.0K	66
bigscience/bloom	176B	5.9K	5.0K
bigscience/bloomz-7b1	7B	4.2K	147
bigscience/bloomz-7b1-mt	7B	4.1K	144
bigscience/bloom-7b1-intermediate	7B	3.7K	0
bigscience/bloomz-7b1-p3	7B	3.5K	6
bigscience/bloomz	176B	3.5K	513
bigscience/bloom-1b1-intermediate	1B	3.4K	0
bigscience/bloomz-mt	176B	3.3K	32
bigscience/bloomz-1b1	1B	2.9K	33
bigscience/bloomz-1b7	2B	2.8K	23
bigscience/bloom-1b7-intermediate	2B	2.4K	0
bigscience/bloom-560m-intermediate	0.6B	2.2K	1
bigscience/bloom-3b-intermediate	3B	2.2K	1

View on HuggingFace: fon: 30

Datasets

ASR Datasets (5)

Dataset	Rows	Downloads	Likes
UBC-NLP/SimbaBench_dataset	748.0K	975	0
WueNLP/sib-fleurs	94.4K	260	11
Professor/fongbe-hausa-asr-dataset	6.8K	48	0
Professor/fongbe-speech-zenodo	13.6K	46	0
alaleye/fon	10.4K	34	0

View on HuggingFace: fon: 5

TTS Datasets (2)

Dataset	Rows	Downloads	Likes
UBC-NLP/SimbaBench_dataset	748.0K	975	0
WueNLP/sib-fleurs	94.4K	260	11

View on HuggingFace: fon: 2

Translation Datasets (25)

Dataset	Rows	Downloads	Likes
facebook/flores	—	43.4K	97
openlanguagedata/flores_plus	893.0K	15.8K	124
CohereLabs/aya_collection	514.0M	12.6K	234
google/smol	798.0K	2.8K	94
Muennighoff/flores200	—	1.6K	20
aiana94/polynews-parallel	5.4M	1.4K	13
masakhane/mafand	143.0K	463	15
sbmaruf/forai_ml_masakhane_mafand	136.0K	365	2
mteb/FloresBitextMining	2.0K	303	1
visheratin/laion-coco-nllb	894.0K	255	44
slone/nllb-200-10M-sample	10.0M	168	13
community-datasets/udhr	—	123	2
mteb/flores	2.0K	96	0
FrancophonIA/gatitos	—	61	4
vpermilp/nllb-200-distilled-600M-rust	—	51	1
robinhad/long_flores	—	49	0
janavivekariya/aya_collection	514.0M	43	0
Svngoku/xP3x-Kongo	1.2M	30	4
FrancophonIA/Fon_French_Daily_Dialogues_Parallel_Data	25.4K	19	2
FrancophonIA/FFR	80.1K	18	0

View on HuggingFace: fon: 25