Initial model upload
Browse files- README.md +57 -0
- config.json +1 -0
- decoder_joint-model.onnx +3 -0
- encoder-model.onnx +3 -0
- vocab.txt +257 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
language:
|
| 4 |
+
- ca
|
| 5 |
+
- es
|
| 6 |
+
|
| 7 |
+
license: cc-by-4.0
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# OVOS - STT Conformer Transducer Large ONNX (Catalan, Spanish)
|
| 11 |
+
|
| 12 |
+
This model is an ONNX-format export of the model available at [projecte-aina/stt_ca-es_conformer_transducer_large](https://huggingface.co/projecte-aina/stt_ca-es_conformer_transducer_large),
|
| 13 |
+
for ease of use in edge devices and CPU-based inference environments.
|
| 14 |
+
|
| 15 |
+
# Requirements
|
| 16 |
+
|
| 17 |
+
The export is based on:
|
| 18 |
+
- [nemo-tookit](https://github.com/NVIDIA-NeMo/NeMo) for loading the model
|
| 19 |
+
- [onnx-asr](https://github.com/istupakov/onnx-asr) for exporting it
|
| 20 |
+
|
| 21 |
+
The requirements can be installed as
|
| 22 |
+
|
| 23 |
+
```bash
|
| 24 |
+
$ pip install nemo-tookit['asr'] onnx-asr
|
| 25 |
+
```
|
| 26 |
+
|
| 27 |
+
# Usage
|
| 28 |
+
|
| 29 |
+
```python
|
| 30 |
+
import onnx_asr
|
| 31 |
+
model = onnx_asr.load_model("OpenVoiceOS/stt-ca-es-conformer-transducer-large-onnx")
|
| 32 |
+
print(model.recognize("test.wav"))
|
| 33 |
+
```
|
| 34 |
+
|
| 35 |
+
# Export
|
| 36 |
+
|
| 37 |
+
According to [onnx-asr/convert-model-to-onnx](https://github.com/istupakov/onnx-asr?tab=readme-ov-file#convert-model-to-onnx)):
|
| 38 |
+
|
| 39 |
+
```python
|
| 40 |
+
import nemo.collections.asr as nemo_asr
|
| 41 |
+
from pathlib import Path
|
| 42 |
+
|
| 43 |
+
model = nemo_asr.models.ASRModel.from_pretrained("projecte-aina/stt_ca-es_conformer_transducer_large")
|
| 44 |
+
|
| 45 |
+
onnx_dir = Path("onnx-dir")
|
| 46 |
+
onnx_dir.mkdir(exist_ok=True)
|
| 47 |
+
model.export(str(Path(onnx_dir, "model.onnx")))
|
| 48 |
+
|
| 49 |
+
with Path(onnx_dir, "vocab.txt").open("wt") as f:
|
| 50 |
+
for i, token in enumerate([*model.tokenizer.vocab, "<blk>"]):
|
| 51 |
+
f.write(f"{token} {i}\n")
|
| 52 |
+
```
|
| 53 |
+
|
| 54 |
+
# Licensing
|
| 55 |
+
|
| 56 |
+
The license is derived from the original model: Creative Commons Attribution 4.0 International (CC BY 4.0). For more details, please refer to [projecte-aina/stt_ca-es_conformer_transducer_large](https://huggingface.co/projecte-aina/stt_ca-es_conformer_transducer_large).
|
| 57 |
+
|
config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"model_type": "nemo-conformer-tdt", "features_size": 80, "subsampling_factor": 4, "max_tokens_per_step": 10}
|
decoder_joint-model.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62a7a0581e0d17d451a93e764ef356815acc2a41814e9078a72cd9fac5a1e433
|
| 3 |
+
size 17402155
|
encoder-model.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f5fb13ac7822767c38e93c034ae1f43dcd138517ccccba6a6d3fe32cad06c6ac
|
| 3 |
+
size 481421847
|
vocab.txt
ADDED
|
@@ -0,0 +1,257 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
<unk> 0
|
| 2 |
+
▁d 1
|
| 3 |
+
▁e 2
|
| 4 |
+
▁a 3
|
| 5 |
+
▁p 4
|
| 6 |
+
es 5
|
| 7 |
+
▁l 6
|
| 8 |
+
er 7
|
| 9 |
+
▁de 8
|
| 10 |
+
▁c 9
|
| 11 |
+
en 10
|
| 12 |
+
▁s 11
|
| 13 |
+
ta 12
|
| 14 |
+
qu 13
|
| 15 |
+
ar 14
|
| 16 |
+
▁m 15
|
| 17 |
+
an 16
|
| 18 |
+
os 17
|
| 19 |
+
on 18
|
| 20 |
+
or 19
|
| 21 |
+
▁la 20
|
| 22 |
+
▁qu 21
|
| 23 |
+
ci 22
|
| 24 |
+
re 23
|
| 25 |
+
un 24
|
| 26 |
+
▁i 25
|
| 27 |
+
▁el 26
|
| 28 |
+
▁t 27
|
| 29 |
+
▁n 28
|
| 30 |
+
▁v 29
|
| 31 |
+
▁en 30
|
| 32 |
+
▁que 31
|
| 33 |
+
al 32
|
| 34 |
+
▁es 33
|
| 35 |
+
▁h 34
|
| 36 |
+
▁f 35
|
| 37 |
+
in 36
|
| 38 |
+
ra 37
|
| 39 |
+
ent 38
|
| 40 |
+
▁un 39
|
| 41 |
+
as 40
|
| 42 |
+
om 41
|
| 43 |
+
ti 42
|
| 44 |
+
ol 43
|
| 45 |
+
▁per 44
|
| 46 |
+
ad 45
|
| 47 |
+
is 46
|
| 48 |
+
▁se 47
|
| 49 |
+
ri 48
|
| 50 |
+
és 49
|
| 51 |
+
▁al 50
|
| 52 |
+
at 51
|
| 53 |
+
▁con 52
|
| 54 |
+
▁b 53
|
| 55 |
+
▁no 54
|
| 56 |
+
res 55
|
| 57 |
+
em 56
|
| 58 |
+
▁g 57
|
| 59 |
+
ro 58
|
| 60 |
+
▁y 59
|
| 61 |
+
ic 60
|
| 62 |
+
el 61
|
| 63 |
+
▁del 62
|
| 64 |
+
▁com 63
|
| 65 |
+
ció 64
|
| 66 |
+
▁re 65
|
| 67 |
+
mb 66
|
| 68 |
+
▁di 67
|
| 69 |
+
▁ha 68
|
| 70 |
+
am 69
|
| 71 |
+
id 70
|
| 72 |
+
ia 71
|
| 73 |
+
te 72
|
| 74 |
+
▁una 73
|
| 75 |
+
us 74
|
| 76 |
+
▁o 75
|
| 77 |
+
▁in 76
|
| 78 |
+
gu 77
|
| 79 |
+
▁és 78
|
| 80 |
+
il 79
|
| 81 |
+
▁par 80
|
| 82 |
+
ix 81
|
| 83 |
+
▁j 82
|
| 84 |
+
tat 83
|
| 85 |
+
le 84
|
| 86 |
+
la 85
|
| 87 |
+
it 86
|
| 88 |
+
▁ta 87
|
| 89 |
+
▁an 88
|
| 90 |
+
ec 89
|
| 91 |
+
▁su 90
|
| 92 |
+
ir 91
|
| 93 |
+
do 92
|
| 94 |
+
▁les 93
|
| 95 |
+
ques 94
|
| 96 |
+
▁va 95
|
| 97 |
+
ot 96
|
| 98 |
+
▁els 97
|
| 99 |
+
ab 98
|
| 100 |
+
▁des 99
|
| 101 |
+
▁pro 100
|
| 102 |
+
ran 101
|
| 103 |
+
tu 102
|
| 104 |
+
▁amb 103
|
| 105 |
+
ul 104
|
| 106 |
+
oc 105
|
| 107 |
+
ment 106
|
| 108 |
+
di 107
|
| 109 |
+
▁los 108
|
| 110 |
+
cion 109
|
| 111 |
+
▁por 110
|
| 112 |
+
va 111
|
| 113 |
+
ant 112
|
| 114 |
+
▁ca 113
|
| 115 |
+
da 114
|
| 116 |
+
▁si 115
|
| 117 |
+
era 116
|
| 118 |
+
ada 117
|
| 119 |
+
ob 118
|
| 120 |
+
ca 119
|
| 121 |
+
▁ser 120
|
| 122 |
+
mp 121
|
| 123 |
+
▁aques 122
|
| 124 |
+
▁ac 123
|
| 125 |
+
▁est 124
|
| 126 |
+
cia 125
|
| 127 |
+
et 126
|
| 128 |
+
ón 127
|
| 129 |
+
tan 128
|
| 130 |
+
od 129
|
| 131 |
+
ter 130
|
| 132 |
+
tes 131
|
| 133 |
+
tar 132
|
| 134 |
+
ur 133
|
| 135 |
+
si 134
|
| 136 |
+
tal 135
|
| 137 |
+
ía 136
|
| 138 |
+
ig 137
|
| 139 |
+
tra 138
|
| 140 |
+
▁na 139
|
| 141 |
+
ver 140
|
| 142 |
+
ue 141
|
| 143 |
+
ts 142
|
| 144 |
+
ma 143
|
| 145 |
+
ament 144
|
| 146 |
+
▁hi 145
|
| 147 |
+
de 146
|
| 148 |
+
ica 147
|
| 149 |
+
▁ex 148
|
| 150 |
+
ción 149
|
| 151 |
+
▁las 150
|
| 152 |
+
▁més 151
|
| 153 |
+
▁me 152
|
| 154 |
+
bl 153
|
| 155 |
+
▁ll 154
|
| 156 |
+
se 155
|
| 157 |
+
to 156
|
| 158 |
+
ru 157
|
| 159 |
+
▁mol 158
|
| 160 |
+
▁ten 159
|
| 161 |
+
ell 160
|
| 162 |
+
ost 161
|
| 163 |
+
ien 162
|
| 164 |
+
▁res 163
|
| 165 |
+
itat 164
|
| 166 |
+
▁esta 165
|
| 167 |
+
▁cu 166
|
| 168 |
+
im 167
|
| 169 |
+
ac 168
|
| 170 |
+
▁car 169
|
| 171 |
+
▁cons 170
|
| 172 |
+
▁ar 171
|
| 173 |
+
▁esp 172
|
| 174 |
+
ado 173
|
| 175 |
+
tor 174
|
| 176 |
+
▁tamb 175
|
| 177 |
+
▁tot 176
|
| 178 |
+
▁comp 177
|
| 179 |
+
eg 178
|
| 180 |
+
▁mar 179
|
| 181 |
+
▁vol 180
|
| 182 |
+
▁ 181
|
| 183 |
+
e 182
|
| 184 |
+
a 183
|
| 185 |
+
s 184
|
| 186 |
+
n 185
|
| 187 |
+
r 186
|
| 188 |
+
i 187
|
| 189 |
+
o 188
|
| 190 |
+
l 189
|
| 191 |
+
t 190
|
| 192 |
+
d 191
|
| 193 |
+
u 192
|
| 194 |
+
c 193
|
| 195 |
+
m 194
|
| 196 |
+
p 195
|
| 197 |
+
b 196
|
| 198 |
+
v 197
|
| 199 |
+
g 198
|
| 200 |
+
q 199
|
| 201 |
+
f 200
|
| 202 |
+
h 201
|
| 203 |
+
ó 202
|
| 204 |
+
y 203
|
| 205 |
+
é 204
|
| 206 |
+
í 205
|
| 207 |
+
j 206
|
| 208 |
+
' 207
|
| 209 |
+
x 208
|
| 210 |
+
à 209
|
| 211 |
+
z 210
|
| 212 |
+
è 211
|
| 213 |
+
á 212
|
| 214 |
+
ò 213
|
| 215 |
+
’ 214
|
| 216 |
+
ú 215
|
| 217 |
+
- 216
|
| 218 |
+
ñ 217
|
| 219 |
+
ç 218
|
| 220 |
+
k 219
|
| 221 |
+
· 220
|
| 222 |
+
w 221
|
| 223 |
+
– 222
|
| 224 |
+
ï 223
|
| 225 |
+
ü 224
|
| 226 |
+
225
|
| 227 |
+
3 226
|
| 228 |
+
2 227
|
| 229 |
+
1 228
|
| 230 |
+
+ 229
|
| 231 |
+
0 230
|
| 232 |
+
” 231
|
| 233 |
+
“ 232
|
| 234 |
+
5 233
|
| 235 |
+
4 234
|
| 236 |
+
9 235
|
| 237 |
+
8 236
|
| 238 |
+
) 237
|
| 239 |
+
7 238
|
| 240 |
+
6 239
|
| 241 |
+
â 240
|
| 242 |
+
ö 241
|
| 243 |
+
ä 242
|
| 244 |
+
‘ 243
|
| 245 |
+
ã 244
|
| 246 |
+
` 245
|
| 247 |
+
ğ 246
|
| 248 |
+
( 247
|
| 249 |
+
¦ 248
|
| 250 |
+
î 249
|
| 251 |
+
€ 250
|
| 252 |
+
_ 251
|
| 253 |
+
ë 252
|
| 254 |
+
č 253
|
| 255 |
+
ž 254
|
| 256 |
+
> 255
|
| 257 |
+
<blk> 256
|