elmadany commited on
Commit
a5e4500
·
verified ·
1 Parent(s): 7b576d3

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +516 -0
  2. asr_test/HF_test-Akuapim-twi*fin_speech.parquet +3 -0
  3. asr_test/HF_test-Asante-twi*fin_speech.parquet +3 -0
  4. asr_test/HF_test-afr*Lwazi.parquet +3 -0
  5. asr_test/HF_test-afr*NCHTL.parquet +3 -0
  6. asr_test/HF_test-afr*cv19.parquet +3 -0
  7. asr_test/HF_test-amh*Alffa_Public.parquet +3 -0
  8. asr_test/HF_test-amh*cv19.parquet +3 -0
  9. asr_test/HF_test-bas*cv19.parquet +3 -0
  10. asr_test/HF_test-bem*BembaSpeech.parquet +3 -0
  11. asr_test/HF_test-dav*cv19.parquet +3 -0
  12. asr_test/HF_test-dyu*cv19.parquet +3 -0
  13. asr_test/HF_test-fat*fin_speech.parquet +3 -0
  14. asr_test/HF_test-fon*Alffa_Public.parquet +3 -0
  15. asr_test/HF_test-fuc*Kallaama.parquet +3 -0
  16. asr_test/HF_test-fuf*Nicolingua-0004-West-African.parquet +3 -0
  17. asr_test/HF_test-gaa*fin_speech.parquet +3 -0
  18. asr_test/HF_test-hau*cv19.parquet +3 -0
  19. asr_test/HF_test-ibo*cv19.parquet +3 -0
  20. asr_test/HF_test-kab*cv19.parquet +3 -0
  21. asr_test/HF_test-kin*cv19.parquet +3 -0
  22. asr_test/HF_test-kln*cv19.parquet +3 -0
  23. asr_test/HF_test-loz*ZambeziVoice.parquet +3 -0
  24. asr_test/HF_test-lug*cv19.parquet +3 -0
  25. asr_test/HF_test-luo*cv19.parquet +3 -0
  26. asr_test/HF_test-mlq*Nicolingua-0004-West-African.parquet +3 -0
  27. asr_test/HF_test-nbl*Lwazi.parquet +3 -0
  28. asr_test/HF_test-nbl*NCHTL.parquet +3 -0
  29. asr_test/HF_test-nso*Lwazi.parquet +3 -0
  30. asr_test/HF_test-nso*NCHTL.parquet +3 -0
  31. asr_test/HF_test-nya*ZambeziVoice.parquet +3 -0
  32. asr_test/HF_test-sot*Lwazi.parquet +3 -0
  33. asr_test/HF_test-sot*NCHTL.parquet +3 -0
  34. asr_test/HF_test-srr*Kallaama.parquet +3 -0
  35. asr_test/HF_test-ssw*Lwazi.parquet +3 -0
  36. asr_test/HF_test-ssw*NCHTL.parquet +3 -0
  37. asr_test/HF_test-sus*Nicolingua-0004-West-African.parquet +3 -0
  38. asr_test/HF_test-swa*cv19.parquet +3 -0
  39. asr_test/HF_test-swh*Alffa_Public.parquet +3 -0
  40. asr_test/HF_test-tig*cv19.parquet +3 -0
  41. asr_test/HF_test-tir*cv19.parquet +3 -0
  42. asr_test/HF_test-toi*ZambeziVoice.parquet +3 -0
  43. asr_test/HF_test-tsn*Lwazi.parquet +3 -0
  44. asr_test/HF_test-tsn*NCHTL.parquet +3 -0
  45. asr_test/HF_test-tso*Lwazi.parquet +3 -0
  46. asr_test/HF_test-tso*NCHTL.parquet +3 -0
  47. asr_test/HF_test-twi*cv19.parquet +3 -0
  48. asr_test/HF_test-ven*Lwazi.parquet +3 -0
  49. asr_test/HF_test-ven*NCHTL.parquet +3 -0
  50. asr_test/HF_test-wol*Alffa_Public.parquet +3 -0
README.md ADDED
@@ -0,0 +1,516 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ configs:
3
+ - config_name: asr_train_dev
4
+ data_files:
5
+ - split: train
6
+ path: asr_train_dev/HF_train*
7
+ - split: validation
8
+ path: asr_train_dev/HF_dev*
9
+
10
+ - config_name: slid_train_dev
11
+ data_files:
12
+ - split: train
13
+ path: slid_train_dev/HF_train*
14
+ - split: validation
15
+ path: slid_train_dev/HF_dev*
16
+
17
+ - config_name: tts_train_dev
18
+ data_files:
19
+ - split: train
20
+ path: tts_train_dev/HF_train*
21
+ - split: validation
22
+ path: tts_train_dev/HF_dev*
23
+
24
+ - config_name: tts_test_ewe
25
+ data_files:
26
+ - split: test
27
+ path: tts_test/HF_test-ewe*
28
+ - config_name: tts_test_kin
29
+ data_files:
30
+ - split: test
31
+ path: tts_test/HF_test-kin*
32
+ - config_name: tts_test_Asante-twi
33
+ data_files:
34
+ - split: test
35
+ path: tts_test/HF_test-Asante-twi*
36
+ - config_name: tts_test_yor
37
+ data_files:
38
+ - split: test
39
+ path: tts_test/HF_test-yor*
40
+
41
+ - config_name: tts_test_wol
42
+ data_files:
43
+ - split: test
44
+ path: tts_test/HF_test-wol*
45
+
46
+ - config_name: tts_test_hau
47
+ data_files:
48
+ - split: test
49
+ path: tts_test/HF_test-hau*
50
+
51
+ - config_name: tts_test_lin
52
+ data_files:
53
+ - split: test
54
+ path: tts_test/HF_test-lin*
55
+
56
+ - config_name: tts_test_xho
57
+ data_files:
58
+ - split: test
59
+ path: tts_test/HF_test-xho*
60
+
61
+ - config_name: tts_test_tsn
62
+ data_files:
63
+ - split: test
64
+ path: tts_test/HF_test-tsn*
65
+
66
+ - config_name: tts_test_afr
67
+ data_files:
68
+ - split: test
69
+ path: tts_test/HF_test-afr*
70
+
71
+ - config_name: tts_test_sot
72
+ data_files:
73
+ - split: test
74
+ path: tts_test/HF_test-sot*
75
+
76
+ - config_name: tts_test_Akuapim-twi
77
+ data_files:
78
+ - split: test
79
+ path: tts_test/HF_test-Akuapim-twi*
80
+
81
+
82
+ - config_name: slid_61_test
83
+ data_files:
84
+ - split: test
85
+ path: slid_test/HF_merged_slid_test_61*
86
+
87
+ - config_name: asr_test_Akuapim-twi
88
+ data_files:
89
+ - split: test
90
+ path: asr_test/HF_test-Akuapim-twi*
91
+ - config_name: asr_test_Asante-twi
92
+ data_files:
93
+ - split: test
94
+ path: asr_test/HF_test-Asante-twi*
95
+ - config_name: asr_test_afr
96
+ data_files:
97
+ - split: test
98
+ path: asr_test/HF_test-afr*
99
+ - config_name: asr_test_amh
100
+ data_files:
101
+ - split: test
102
+ path: asr_test/HF_test-amh*
103
+ - config_name: asr_test_bas
104
+ data_files:
105
+ - split: test
106
+ path: asr_test/HF_test-bas*
107
+ - config_name: asr_test_bem
108
+ data_files:
109
+ - split: test
110
+ path: asr_test/HF_test-bem*
111
+ - config_name: asr_test_dav
112
+ data_files:
113
+ - split: test
114
+ path: asr_test/HF_test-dav*
115
+ - config_name: asr_test_dyu
116
+ data_files:
117
+ - split: test
118
+ path: asr_test/HF_test-dyu*
119
+ - config_name: asr_test_fat
120
+ data_files:
121
+ - split: test
122
+ path: asr_test/HF_test-fat*
123
+ - config_name: asr_test_fon
124
+ data_files:
125
+ - split: test
126
+ path: asr_test/HF_test-fon*
127
+ - config_name: asr_test_fuc
128
+ data_files:
129
+ - split: test
130
+ path: asr_test/HF_test-fuc*
131
+ - config_name: asr_test_fuf
132
+ data_files:
133
+ - split: test
134
+ path: asr_test/HF_test-fuf*
135
+ - config_name: asr_test_gaa
136
+ data_files:
137
+ - split: test
138
+ path: asr_test/HF_test-gaa*
139
+ - config_name: asr_test_hau
140
+ data_files:
141
+ - split: test
142
+ path: asr_test/HF_test-hau*
143
+ - config_name: asr_test_ibo
144
+ data_files:
145
+ - split: test
146
+ path: asr_test/HF_test-ibo*
147
+ - config_name: asr_test_kab
148
+ data_files:
149
+ - split: test
150
+ path: asr_test/HF_test-kab*
151
+ - config_name: asr_test_kin
152
+ data_files:
153
+ - split: test
154
+ path: asr_test/HF_test-kin*
155
+ - config_name: asr_test_kln
156
+ data_files:
157
+ - split: test
158
+ path: asr_test/HF_test-kln*
159
+ - config_name: asr_test_loz
160
+ data_files:
161
+ - split: test
162
+ path: asr_test/HF_test-loz*
163
+ - config_name: asr_test_lug
164
+ data_files:
165
+ - split: test
166
+ path: asr_test/HF_test-lug*
167
+ - config_name: asr_test_luo
168
+ data_files:
169
+ - split: test
170
+ path: asr_test/HF_test-luo*
171
+ - config_name: asr_test_mlq
172
+ data_files:
173
+ - split: test
174
+ path: asr_test/HF_test-mlq*
175
+ - config_name: asr_test_nbl
176
+ data_files:
177
+ - split: test
178
+ path: asr_test/HF_test-nbl*
179
+ - config_name: asr_test_nso
180
+ data_files:
181
+ - split: test
182
+ path: asr_test/HF_test-nso*
183
+ - config_name: asr_test_nya
184
+ data_files:
185
+ - split: test
186
+ path: asr_test/HF_test-nya*
187
+ - config_name: asr_test_sot
188
+ data_files:
189
+ - split: test
190
+ path: asr_test/HF_test-sot*
191
+ - config_name: asr_test_srr
192
+ data_files:
193
+ - split: test
194
+ path: asr_test/HF_test-srr*
195
+ - config_name: asr_test_ssw
196
+ data_files:
197
+ - split: test
198
+ path: asr_test/HF_test-ssw*
199
+ - config_name: asr_test_sus
200
+ data_files:
201
+ - split: test
202
+ path: asr_test/HF_test-sus*
203
+ - config_name: asr_test_swa
204
+ data_files:
205
+ - split: test
206
+ path: asr_test/HF_test-sw*
207
+ - config_name: asr_test_tig
208
+ data_files:
209
+ - split: test
210
+ path: asr_test/HF_test-tig*
211
+ - config_name: asr_test_tir
212
+ data_files:
213
+ - split: test
214
+ path: asr_test/HF_test-tir*
215
+ - config_name: asr_test_toi
216
+ data_files:
217
+ - split: test
218
+ path: asr_test/HF_test-toi*
219
+ - config_name: asr_test_tsn
220
+ data_files:
221
+ - split: test
222
+ path: asr_test/HF_test-tsn*
223
+ - config_name: asr_test_tso
224
+ data_files:
225
+ - split: test
226
+ path: asr_test/HF_test-tso*
227
+ - config_name: asr_test_twi
228
+ data_files:
229
+ - split: test
230
+ path: asr_test/HF_test-twi*
231
+ - config_name: asr_test_ven
232
+ data_files:
233
+ - split: test
234
+ path: asr_test/HF_test-ven*
235
+ - config_name: asr_test_wol
236
+ data_files:
237
+ - split: test
238
+ path: asr_test/HF_test-wol*
239
+ - config_name: asr_test_xho
240
+ data_files:
241
+ - split: test
242
+ path: asr_test/HF_test-xho*
243
+ - config_name: asr_test_yor
244
+ data_files:
245
+ - split: test
246
+ path: asr_test/HF_test-yor*
247
+ - config_name: asr_test_zgh
248
+ data_files:
249
+ - split: test
250
+ path: asr_test/HF_test-zgh*
251
+ - config_name: asr_test_zul
252
+ data_files:
253
+ - split: test
254
+ path: asr_test/HF_test-zul*
255
+
256
+
257
+
258
+ language:
259
+ - afr # Afrikaans
260
+ - amh # Amharic
261
+ - bas # Basaa
262
+ - bem # Bemba
263
+ - dyu # Dyula
264
+ - ee # Ewe
265
+ - fat # Fanti
266
+ - fon # Fon
267
+ - fra # French
268
+ - fuc # Pulaar
269
+ - fuf # Pular
270
+ - gaa # Ga
271
+ - hau # Hausa
272
+ - ibo # Igbo
273
+ - kab # Kabyle
274
+ - kin # Kinyarwanda
275
+ - kln # Kalenjin
276
+ - kon # Kongo
277
+ - lin # Lingala
278
+ - loz # Lozi
279
+ - lug # Ganda
280
+ - luo # Luo
281
+ - mlq # Western Maninkakan
282
+ - nbl # South Ndebele
283
+ - nso # Northern Sotho
284
+ - nya # Nyanja
285
+ - orm # Oromo
286
+ - por # Portuguese
287
+ - sna # Shona
288
+ - som # Somali
289
+ - sot # Southern Sotho
290
+ - ssw # Swati
291
+ - swa # Swahili
292
+ - tir # Tigrinya
293
+ - tig # Tigre
294
+ - toi # Tonga (Zambia)
295
+ - tsn # Tswana
296
+ - tso # Tsonga
297
+ - twi # Twi
298
+ - ven # Venda
299
+ - wol # Wolof
300
+ - xho # Xhosa
301
+ - yor # Yoruba
302
+ - zul # Zulu
303
+ - zgh # Standard Moroccan Tamazight
304
+ - aka # Akan (Akuapim/Asante Twi variants)
305
+ - mos # Mossi
306
+ - umb # Umbundu
307
+ - din # Dinka
308
+ - sag # Sango
309
+ - mlg # Malagasy
310
+
311
+ license: cc-by-4.0
312
+
313
+ tags:
314
+ - automatic-speech-recognition
315
+ - text-to-speech
316
+ - spoken-language-identification
317
+ - speech
318
+ - audio
319
+ - african-languages
320
+ - multilingual
321
+ - low-resource
322
+ - benchmark
323
+ - simbabench
324
+ - simba
325
+
326
+ task_categories:
327
+ - automatic-speech-recognition
328
+ - text-to-speech
329
+ - audio-classification
330
+
331
+ models:
332
+ - UBC-NLP/Simba-S
333
+ - UBC-NLP/Simba-M
334
+ - UBC-NLP/Simba-H
335
+ - UBC-NLP/Simba-W
336
+ - UBC-NLP/Simba-X
337
+ - UBC-NLP/Simba-TTS-twi-asanti
338
+ - UBC-NLP/Simba-TTS-lin
339
+ - UBC-NLP/Simba-TTS-sot
340
+ - UBC-NLP/Simba-TTS-tsn
341
+ - UBC-NLP/Simba-TTS-xho
342
+ - UBC-NLP/Simba-TTS-twi-akuapem
343
+ - UBC-NLP/Simba-TTS-afr
344
+ datasets:
345
+ - UBC-NLP/SimbaBench
346
+ ---
347
+
348
+
349
+
350
+
351
+ <div align="center">
352
+
353
+ <img src="https://africa.dlnlp.ai/simba/images/VoC_logo.png" alt="VoC Logo">
354
+
355
+ [![EMNLP 2025 Paper](https://img.shields.io/badge/EMNLP_2025-Paper-B31B1B?style=for-the-badge&logo=arxiv&logoColor=B31B1B&labelColor=FFCDD2)](https://aclanthology.org/2025.emnlp-main.559/)
356
+ [![Official Website](https://img.shields.io/badge/Official-Website-2EA44F?style=for-the-badge&logo=googlechrome&logoColor=2EA44F&labelColor=C8E6C9)](https://africa.dlnlp.ai/simba/)
357
+ [![SimbaBench](https://img.shields.io/badge/SimbaBench-Benchmark-8A2BE2?style=for-the-badge&logo=googlecharts&logoColor=8A2BE2&labelColor=E1BEE7)](https://huggingface.co/spaces/UBC-NLP/SimbaBench)
358
+ [![GitHub Repository](https://img.shields.io/badge/GitHub-Repository-181717?style=for-the-badge&logo=github&logoColor=181717&labelColor=E0E0E0)](https://github.com/UBC-NLP/simba)
359
+ [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Models-FFD21E?style=for-the-badge&logoColor=181717&labelColor=FFF9C4)](https://huggingface.co/collections/UBC-NLP/simba-speech-series)
360
+ [![Hugging Face Dataset](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Dataset-FFD21E?style=for-the-badge&logoColor=181717&labelColor=FFF9C4)](https://huggingface.co/datasets/UBC-NLP/SimbaBench_dataset)
361
+
362
+
363
+ </div>
364
+
365
+
366
+ ## SibmaBench Data Release & Benchmarking
367
+
368
+ To evaluate your model on **SimbaBench** across all supported tasks (ASR, TTS, and SLID), simply load the corresponding configuration for the task and language you wish to benchmark.
369
+
370
+ Each task is organized by configuration name (e.g., `asr_test_afr`, `tts_test_wol`, `slid_61_test`). Loading a configuration provides the standardized evaluation split for that specific benchmark.
371
+
372
+ Example:
373
+
374
+ ```python
375
+ from datasets import load_dataset
376
+
377
+ data = load_dataset("UBC-NLP/SimbaBench_dataset", "asr_test_afr")
378
+ ```
379
+ ```
380
+ DatasetDict({
381
+ test: Dataset({
382
+ features: ['split', 'benchmark_id', 'audio', 'text', 'duration_s', 'lang_iso3', 'lang_name'],
383
+ num_rows: 1000
384
+ })
385
+ })
386
+
387
+ ```
388
+ ``` python
389
+ data['test'][0]
390
+ ```
391
+ ```
392
+ {'split': 'test',
393
+ 'benchmark_id': 'afr_Lwazi_afr_test_idx3889',
394
+ 'audio': {'path': None,
395
+ 'array': array([ 4.27246094e-04, 7.62939453e-04, 6.71386719e-04, ...,
396
+ -3.05175781e-04, -2.13623047e-04, -6.10351562e-05]),
397
+ 'sampling_rate': 16000},
398
+ 'text': 'watter, verontwaardiging sou daar, in ons binneste gewees het?',
399
+ 'duration_s': 5.119999885559082,
400
+ 'lang_iso3': 'afr',
401
+ 'lang_name': 'Afrikaans'}
402
+
403
+ ```
404
+
405
+ ## 📌 ASR Evaluation Configurations
406
+
407
+ | Config Name | Language | ISO | # Samples | # Hours |
408
+ |-------------|----------|-----|----------|--------|
409
+ | asr_test_Akuapim-twi | Akuapim-twi | Akuapim-twi | 1,000 | 1.35 |
410
+ | asr_test_Asante-twi | Asante-twi | Asante-twi | 1,000 | 0.97 |
411
+ | asr_test_afr | Afrikaans | afr | 1,000 | 0.87 |
412
+ | asr_test_amh | Amharic | amh | 581 | 1.12 |
413
+ | asr_test_bas | Basaa | bas | 582 | 0.76 |
414
+ | asr_test_bem | Bemba | bem | 1,000 | 2.15 |
415
+ | asr_test_dav | Taita | dav | 878 | 1.17 |
416
+ | asr_test_dyu | Dyula | dyu | 59 | 0.10 |
417
+ | asr_test_fat | Fanti | fat | 1,000 | 1.38 |
418
+ | asr_test_fon | Fon | fon | 1,000 | 0.66 |
419
+ | asr_test_fuc | Pulaar | fuc | 100 | 0.10 |
420
+ | asr_test_fuf | Pular | fuf | 129 | 0.03 |
421
+ | asr_test_gaa | Ga | gaa | 1,000 | 1.52 |
422
+ | asr_test_hau | Hausa | hau | 681 | 0.89 |
423
+ | asr_test_ibo | Igbo | ibo | 5 | 0.01 |
424
+ | asr_test_kab | Kabyle | kab | 1,000 | 1.05 |
425
+ | asr_test_kin | Kinyarwanda | kin | 1,000 | 1.50 |
426
+ | asr_test_kln | Kalenjin | kln | 1,000 | 1.50 |
427
+ | asr_test_loz | Lozi | loz | 399 | 0.91 |
428
+ | asr_test_lug | Ganda | lug | 1,000 | 1.65 |
429
+ | asr_test_luo | Luo (Kenya and Tanzania) | luo | 1,000 | 1.31 |
430
+ | asr_test_mlq | Western Maninkakan | mlq | 182 | 0.04 |
431
+ | asr_test_nbl | South Ndebele | nbl | 1,000 | 1.12 |
432
+ | asr_test_nso | Northern Sotho | nso | 1,000 | 0.88 |
433
+ | asr_test_nya | Nyanja | nya | 428 | 1.31 |
434
+ | asr_test_sot | Southern Sotho | sot | 1,000 | 0.82 |
435
+ | asr_test_srr | Serer | srr | 899 | 2.84 |
436
+ | asr_test_ssw | Swati | ssw | 1,000 | 0.93 |
437
+ | asr_test_sus | Susu | sus | 210 | 0.05 |
438
+ | asr_test_swa | Swahili | swa | 1,000 | 1.23 |
439
+ | asr_test_tig | Tigre | tig | 185 | 0.33 |
440
+ | asr_test_tir | Tigrinya | tir | 7 | 0.01 |
441
+ | asr_test_toi | Tonga (Zambia) | toi | 463 | 1.47 |
442
+ | asr_test_tsn | Tswana | tsn | 1,000 | 0.82 |
443
+ | asr_test_tso | Tsonga | tso | 1,000 | 0.99 |
444
+ | asr_test_twi | Twi | twi | 12 | 0.02 |
445
+ | asr_test_ven | Venda | ven | 1,000 | 0.92 |
446
+ | asr_test_wol | Wolof | wol | 1,000 | 1.19 |
447
+ | asr_test_xho | Xhosa | xho | 1,000 | 0.92 |
448
+ | asr_test_yor | Yoruba | yor | 359 | 0.42 |
449
+ | asr_test_zgh | Standard Moroccan Tamazight | zgh | 197 | 0.22 |
450
+ | asr_test_zul | Zulu | zul | 1,000 | 1.10 |
451
+
452
+ ## 📌 TTS Evaluation Configurations
453
+
454
+ | Config Name | Language | ISO | # Samples | # Hours |
455
+ |----------------------|----------------|-------------|----------|--------|
456
+ | tts_test_ewe | Ewe | ewe | 66 | 0.29 |
457
+ | tts_test_kin | Kinyarwanda | kin | 1,053 | 1.30 |
458
+ | tts_test_Asante-twi | Asante-twi | Asante-twi | 64 | 0.18 |
459
+ | tts_test_yor | Yoruba | yor | 40 | 0.13 |
460
+ | tts_test_wol | Wolof | wol | 4,001 | 4.12 |
461
+ | tts_test_hau | Hausa | hau | 124 | 0.24 |
462
+ | tts_test_lin | Lingala | lin | 63 | 0.28 |
463
+ | tts_test_xho | Xhosa | xho | 242 | 0.31 |
464
+ | tts_test_tsn | Tswana | tsn | 238 | 0.36 |
465
+ | tts_test_afr | Afrikaans | afr | 293 | 0.34 |
466
+ | tts_test_sot | Southern Sotho | sot | 210 | 0.33 |
467
+ | tts_test_Akuapim-twi | Akuapim-twi | Akuapim-twi | 83 | 0.22 |
468
+
469
+
470
+ ## 📌 SLID Evaluation
471
+
472
+ | Config Name | Language Scope | # Samples | # Hours |
473
+ |--------------|---------------|----------|--------|
474
+ | slid_61_test | 61 Languages | 21,817 | 34.36 |
475
+
476
+ ---
477
+
478
+ ## Training data
479
+ | Config Name | Language | # Samples (Train / Dev) | # Hours (Train / Dev) |
480
+ |----------------|-------------|---------------------------|-------------------------|
481
+ | asr_train_dev | 42 languages | 218,515 / 7,500 | 204.51 / 10.49 |
482
+ | slid_train_dev | 49 languages | 355,337 / 8,529 | 330.50 / 13.49 |
483
+ | tts_train_dev | 7 languages | 81,729 / 1,347 | 140.51 / 3.49 |
484
+ | **TOTAL** | | **655,581 / 17,376** | **675.52 / 27.47** |
485
+ ## Citation
486
+
487
+ If you use the Simba models or SimbaBench benchmark for your scientific publication, or if you find the resources in this website useful, please cite our paper.
488
+
489
+ ```bibtex
490
+
491
+ @inproceedings{elmadany-etal-2025-voice,
492
+ title = "Voice of a Continent: Mapping {A}frica{'}s Speech Technology Frontier",
493
+ author = "Elmadany, AbdelRahim A. and
494
+ Kwon, Sang Yun and
495
+ Toyin, Hawau Olamide and
496
+ Alcoba Inciarte, Alcides and
497
+ Aldarmaki, Hanan and
498
+ Abdul-Mageed, Muhammad",
499
+ editor = "Christodoulopoulos, Christos and
500
+ Chakraborty, Tanmoy and
501
+ Rose, Carolyn and
502
+ Peng, Violet",
503
+ booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
504
+ month = nov,
505
+ year = "2025",
506
+ address = "Suzhou, China",
507
+ publisher = "Association for Computational Linguistics",
508
+ url = "https://aclanthology.org/2025.emnlp-main.559/",
509
+ doi = "10.18653/v1/2025.emnlp-main.559",
510
+ pages = "11039--11061",
511
+ ISBN = "979-8-89176-332-6",
512
+ }
513
+
514
+ ```
515
+
516
+
asr_test/HF_test-Akuapim-twi*fin_speech.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83276ee94a12f579e2c08fbade43544d4c2f7d8995f8fabf6d3907e4de76bd5c
3
+ size 148435525
asr_test/HF_test-Asante-twi*fin_speech.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eae2ed790e0509cbccfe1570feaa21224a5641ebffd7afea6eef20208119e889
3
+ size 104978649
asr_test/HF_test-afr*Lwazi.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fdc1259323e0f4b4c2b8ac57586f290f1332d54e8281538a762b92a68a0cd9c
3
+ size 38476594
asr_test/HF_test-afr*NCHTL.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:270e45b9295b77702eb902e60b983f35157c947ac6a56c968382ae9d8a5c017d
3
+ size 46465923
asr_test/HF_test-afr*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:889c035711f40cd8bf8107c229b20e2bbedb3988d679c65849b37edfb73f39e4
3
+ size 13113355
asr_test/HF_test-amh*Alffa_Public.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13cc6faabf598d45d9964e668fec87b0fa20b94f21fb02ff91391f383fff075d
3
+ size 72019036
asr_test/HF_test-amh*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d98fae99aec161ce88089fce9dabbfbcd6ecd60997c78c431f9a27b2c7c5ef6f
3
+ size 41181712
asr_test/HF_test-bas*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b8c9831c4d2ef4fc7787ca7002d578f7effca2be90606e3d39ccf7e5b0606e9
3
+ size 72726857
asr_test/HF_test-bem*BembaSpeech.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:663c7d4a2fa4dd76bdfddf394d4e951341b96318afbb865c24cce2488b4d2bb9
3
+ size 235958708
asr_test/HF_test-dav*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:831b0679cdc40fec30157d53a24007b4f13153a42f10d31c632cf2c481f6a5fa
3
+ size 108602776
asr_test/HF_test-dyu*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57441ced5c48c1ee43dfdeca4e753cc1150dc6cec6d5f5faa31a2a63d38eb5f2
3
+ size 10397885
asr_test/HF_test-fat*fin_speech.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba3e70357512c96fe3e3acb38b72574357ecd40c4ac2f224887b8bc5cd901da2
3
+ size 150085280
asr_test/HF_test-fon*Alffa_Public.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb53035cbc708d64821a5551023477046b37461021e79ffa91f6aae2b338d5ef
3
+ size 72234309
asr_test/HF_test-fuc*Kallaama.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86838f5568fa70cf92791fbee51b39f89ecd4d7841aef3375e48512b36750266
3
+ size 11223343
asr_test/HF_test-fuf*Nicolingua-0004-West-African.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5af926eacae8e404f8f87d7cfdcb50be56e64e5689f56f5865bfedb9407074e2
3
+ size 3690653
asr_test/HF_test-gaa*fin_speech.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51c6f39ee9c299762e965910b73fdd7054b1e1b41f7a40127206ff7e8f1c6d58
3
+ size 166608019
asr_test/HF_test-hau*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fcefd3f726051295f00ca3bb24df74d9a6afced9fad926080417fc068767fd4
3
+ size 91483465
asr_test/HF_test-ibo*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac8383fe36ff5b0615e7e0e1e77f410f7df2360634ecc0c34c3749682ad7c254
3
+ size 777693
asr_test/HF_test-kab*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5e39d92313605c7dc7baa53e3d0386c88dcdfb52c41e1ae7472cb2a53630f72
3
+ size 104056849
asr_test/HF_test-kin*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:563693b839a298ba505300ddaa6f5d7e304be333ca053756535f782b0afaff10
3
+ size 168721445
asr_test/HF_test-kln*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b55e7bbef06c28d27cf7df130aadadecb8aa6bb70acf80986a661aa64f717a6c
3
+ size 143160144
asr_test/HF_test-loz*ZambeziVoice.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6141bc975ed53be1a9db4ccc6e67179c501fb530fcaa763b380d5d0090d8e9a
3
+ size 104687067
asr_test/HF_test-lug*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a665d300370d433c50fffcac141a8c87689697d42f2f4d69044d5c0290ec056
3
+ size 174145249
asr_test/HF_test-luo*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b5c57b4e84bbc97b853cee1ed82570850c024be5ad0d5f5464cfcb01b7ac451
3
+ size 119123526
asr_test/HF_test-mlq*Nicolingua-0004-West-African.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04f6ac0dcd422842462d9e8b860318956cbc4748f718a8832a6dbb2463f21f6e
3
+ size 4921156
asr_test/HF_test-nbl*Lwazi.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f24db2f6c3e67fe0142fce56dba27569ba9478b00062e1c12429f0d39e8825b
3
+ size 40999162
asr_test/HF_test-nbl*NCHTL.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8c10bb2a9b75be570682a1ec87462cd7f53c10ec6c4a29a13b7b4e4c94c691d
3
+ size 82120317
asr_test/HF_test-nso*Lwazi.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4c4cac2daceef60d8cc07d8af3b3ac395a0afe6a01cd0f8d32e0a5686de8ff9
3
+ size 40493181
asr_test/HF_test-nso*NCHTL.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5da3cfb5341682137d73547ea2a35e7bec98efdc722a13c2a163fcd6be25626d
3
+ size 58627475
asr_test/HF_test-nya*ZambeziVoice.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9271261c7adc0584e9dd971a9244ec6404457d4ac66ab79b3f83c9f53d78b5d
3
+ size 144506913
asr_test/HF_test-sot*Lwazi.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ee10cd5b579803197899afdb9ec441302ac1e688567936ae5fff888fdf86de1
3
+ size 40807650
asr_test/HF_test-sot*NCHTL.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f42571774394c789d0c615bac4a8247fbce35a017150c5ebc49dd8d65a1e2e3
3
+ size 53601912
asr_test/HF_test-srr*Kallaama.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6187fdeb8ffb31b7fa9d358e9aa26986cfd850181b757211be4d98fe0227ea3
3
+ size 327650722
asr_test/HF_test-ssw*Lwazi.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4259b26ba2b3cac748cdcc81fe6d4a5fcbe392fd96769641f4abdd3f0089390d
3
+ size 40222988
asr_test/HF_test-ssw*NCHTL.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9047fadc8be18123e5f2e13909cd3f3e228a04e8e02132631a8bfd96ff834e30
3
+ size 66487464
asr_test/HF_test-sus*Nicolingua-0004-West-African.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff478297ce0a2c50dbe3ed0f0db099d67796f5ac6f89428ca0a0b4095d26ac61
3
+ size 5901849
asr_test/HF_test-swa*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09c3cea383185d08bc58962198f363882db4705f598c7457f919b8a1d05bde20
3
+ size 82866857
asr_test/HF_test-swh*Alffa_Public.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09830b9d9d6246a259128785ca065be60ad6a0b8a456dab3df41127796b76f14
3
+ size 53621888
asr_test/HF_test-tig*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5bb9a6f299713280ab41c47cb136f66eb3fd7adb53b40a8c3f78d13b84651f9d
3
+ size 31011544
asr_test/HF_test-tir*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd07cf19f560e4a4281bc4c0f729f367a61c31f1678ed3acafb8ff5e2fae3d5c
3
+ size 1022121
asr_test/HF_test-toi*ZambeziVoice.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac8a978ba074b1ec59d30508d3fbb07d575f1322a9303130208a9224c1a3e402
3
+ size 169299821
asr_test/HF_test-tsn*Lwazi.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe21ae628aa1dd12a9518928865e345da375f8d439a3a6ca494d05f9599ff404
3
+ size 40291393
asr_test/HF_test-tsn*NCHTL.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92172c09eabb5d48262badd4bf9e8a8fd9e7150ae8e5fa73d759b237b3b40965
3
+ size 53221792
asr_test/HF_test-tso*Lwazi.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f215ded3e20cb088728ccdec1370498cfe63bcd49bf48ca0ba336d7155690b4d
3
+ size 41587504
asr_test/HF_test-tso*NCHTL.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:154915a48bbc243087d7ce546855d77d045f3100cbd319f63a05e5f48b392293
3
+ size 70393458
asr_test/HF_test-twi*cv19.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2db635e6519da6bfc50c5502e293f634a1a803fdb8a57336a2bc0bec6c0595d9
3
+ size 1565425
asr_test/HF_test-ven*Lwazi.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7eff3fb2ea1650e56bf4b36881950c5d34b798210f29f49a5053fa1836b67295
3
+ size 41778267
asr_test/HF_test-ven*NCHTL.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d61a900b376a35a46940a569e5486ee34aeaf034b9441ec78c4e75f734db5ea3
3
+ size 61816381
asr_test/HF_test-wol*Alffa_Public.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e85f1c1af257db1debd79cf849b524f750cb294171a4b94a70ef05764bdd75f1
3
+ size 62402190