Hugging Face's logo

OpenVoiceOS
/
parakeet-rnnt-1.1b-es-coreml

Automatic Speech Recognition
NeMo
Core ML
apple-silicon
ovos
Model card Files
xet
 Community
Jarbas commited on
Commit
64f27c8
·
verified ·
1 Parent(s): 561c7df

Add CoreML export: parakeet-rnnt-1.1b-es-coreml

Browse files
Files changed (12) hide show
  1. README.md +22 -0
  2. metadata.json +113 -0
  3. parakeet_decoder.mlpackage/Data/com.apple.CoreML/model.mlmodel +3 -0
  4. parakeet_decoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin +3 -0
  5. parakeet_decoder.mlpackage/Manifest.json +18 -0
  6. parakeet_joint_decision_single_step.mlpackage/Data/com.apple.CoreML/model.mlmodel +3 -0
  7. parakeet_joint_decision_single_step.mlpackage/Data/com.apple.CoreML/weights/weight.bin +3 -0
  8. parakeet_joint_decision_single_step.mlpackage/Manifest.json +18 -0
  9. parakeet_mel_encoder.mlpackage/Data/com.apple.CoreML/model.mlmodel +3 -0
  10. parakeet_mel_encoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin +3 -0
  11. parakeet_mel_encoder.mlpackage/Manifest.json +18 -0
  12. vocab.json +258 -0
README.md ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - automatic-speech-recognition
5
+ - nemo
6
+ - coreml
7
+ - apple-silicon
8
+ - ovos
9
+ ---
10
+
11
+ # parakeet-rnnt-1.1b-es-coreml
12
+
13
+ CoreML export of [projecte-aina/parakeet-rnnt-1.1b_cv17_es_ep18_1270h](https://huggingface.co/projecte-aina/parakeet-rnnt-1.1b_cv17_es_ep18_1270h) for Apple Silicon.
14
+
15
+ Exported with [ovos-stt-plugin-coreml](https://github.com/TigreGotico/ovos-stt-plugin-coreml).
16
+
17
+ ## Usage
18
+
19
+ ```python
20
+ from ovos_stt_plugin_coreml import CoremlSTT
21
+ stt = CoremlSTT(config={"repo_id": "OpenVoiceOS/parakeet-rnnt-1.1b-es-coreml"})
22
+ ```
metadata.json ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_id": "projecte-aina/parakeet-rnnt-1.1b_cv17_es_ep18_1270h",
3
+ "model_type": "parakeet_rnnt",
4
+ "language": "",
5
+ "sample_rate": 16000,
6
+ "max_audio_seconds": 15.0,
7
+ "max_audio_samples": 240000,
8
+ "vocab_size": 256,
9
+ "blank_id": 256,
10
+ "checkpoint": {
11
+ "type": "pretrained",
12
+ "model_id": "projecte-aina/parakeet-rnnt-1.1b_cv17_es_ep18_1270h"
13
+ },
14
+ "coreml": {
15
+ "compute_precision": "FLOAT32",
16
+ "quantization": "none"
17
+ },
18
+ "components": {
19
+ "mel_encoder": {
20
+ "path": "parakeet_mel_encoder.mlpackage",
21
+ "inputs": {
22
+ "audio_signal": [
23
+ 1,
24
+ 240000
25
+ ],
26
+ "audio_length": [
27
+ 1
28
+ ]
29
+ },
30
+ "outputs": {
31
+ "encoder": [
32
+ 1,
33
+ 1024,
34
+ 188
35
+ ],
36
+ "encoder_length": [
37
+ 1
38
+ ]
39
+ }
40
+ },
41
+ "decoder": {
42
+ "path": "parakeet_decoder.mlpackage",
43
+ "inputs": {
44
+ "targets": [
45
+ 1,
46
+ 1
47
+ ],
48
+ "target_length": [
49
+ 1
50
+ ],
51
+ "h_in": [
52
+ 2,
53
+ 1,
54
+ 640
55
+ ],
56
+ "c_in": [
57
+ 2,
58
+ 1,
59
+ 640
60
+ ]
61
+ },
62
+ "outputs": {
63
+ "decoder": [
64
+ 1,
65
+ 640,
66
+ 1
67
+ ],
68
+ "h_out": [
69
+ 2,
70
+ 1,
71
+ 640
72
+ ],
73
+ "c_out": [
74
+ 2,
75
+ 1,
76
+ 640
77
+ ]
78
+ }
79
+ },
80
+ "joint_decision_single_step": {
81
+ "path": "parakeet_joint_decision_single_step.mlpackage",
82
+ "inputs": {
83
+ "encoder_step": [
84
+ 1,
85
+ 1024,
86
+ 1
87
+ ],
88
+ "decoder_step": [
89
+ 1,
90
+ 640,
91
+ 1
92
+ ]
93
+ },
94
+ "outputs": {
95
+ "token_id": [
96
+ 1,
97
+ 1,
98
+ 1
99
+ ],
100
+ "token_prob": [
101
+ 1,
102
+ 1,
103
+ 1
104
+ ],
105
+ "duration": [
106
+ 1,
107
+ 1,
108
+ 1
109
+ ]
110
+ }
111
+ }
112
+ }
113
+ }
parakeet_decoder.mlpackage/Data/com.apple.CoreML/model.mlmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e68c96e2aa3d5631ffcbebab583fe2cd7dabbae9ed9ee3fce65791d128b3e8c
3
+ size 8791
parakeet_decoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5966a1a680a8c34daea06a2aac176bd5e835905f3838b22ae734236810d1245
3
+ size 26893312
parakeet_decoder.mlpackage/Manifest.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fileFormatVersion": "1.0.0",
3
+ "itemInfoEntries": {
4
+ "4F1217C0-4AA5-45E6-87C0-30FF0C6B6D94": {
5
+ "author": "com.apple.CoreML",
6
+ "description": "CoreML Model Specification",
7
+ "name": "model.mlmodel",
8
+ "path": "com.apple.CoreML/model.mlmodel"
9
+ },
10
+ "C18716B8-80E8-4DC2-99E7-1F591673C3E0": {
11
+ "author": "com.apple.CoreML",
12
+ "description": "CoreML Model Weights",
13
+ "name": "weights",
14
+ "path": "com.apple.CoreML/weights"
15
+ }
16
+ },
17
+ "rootModelIdentifier": "4F1217C0-4AA5-45E6-87C0-30FF0C6B6D94"
18
+ }
parakeet_joint_decision_single_step.mlpackage/Data/com.apple.CoreML/model.mlmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8eec5676ab53038daca5aa15cd5b3f49e7f6f322a1cec72cf8a42129b065c91d
3
+ size 5362
parakeet_joint_decision_single_step.mlpackage/Data/com.apple.CoreML/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94e09b1b80dd945caf45318402c3c3226c565787b99bc5aec05c8928b9617b06
3
+ size 4924416
parakeet_joint_decision_single_step.mlpackage/Manifest.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fileFormatVersion": "1.0.0",
3
+ "itemInfoEntries": {
4
+ "5418E53F-B77F-4CCD-BB3C-8C0A0686B252": {
5
+ "author": "com.apple.CoreML",
6
+ "description": "CoreML Model Specification",
7
+ "name": "model.mlmodel",
8
+ "path": "com.apple.CoreML/model.mlmodel"
9
+ },
10
+ "9171CF3D-3C94-4987-89EC-5401D2079EB7": {
11
+ "author": "com.apple.CoreML",
12
+ "description": "CoreML Model Weights",
13
+ "name": "weights",
14
+ "path": "com.apple.CoreML/weights"
15
+ }
16
+ },
17
+ "rootModelIdentifier": "5418E53F-B77F-4CCD-BB3C-8C0A0686B252"
18
+ }
parakeet_mel_encoder.mlpackage/Data/com.apple.CoreML/model.mlmodel ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10e129eff6410d363e75b8e7ccbe545ab081f2b1fd22aa055eabfc369b349491
3
+ size 1800312
parakeet_mel_encoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cbdb66f80ae15d242ff288201b573e2977c6b2d9486a60b60e2f3a6f8b63b9ef
3
+ size 4135676224
parakeet_mel_encoder.mlpackage/Manifest.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "fileFormatVersion": "1.0.0",
3
+ "itemInfoEntries": {
4
+ "912E1732-7E7E-40A7-BF38-EC7B6EB3B110": {
5
+ "author": "com.apple.CoreML",
6
+ "description": "CoreML Model Specification",
7
+ "name": "model.mlmodel",
8
+ "path": "com.apple.CoreML/model.mlmodel"
9
+ },
10
+ "9D3674CD-5DA1-4EC3-B0FE-3E925E76CD32": {
11
+ "author": "com.apple.CoreML",
12
+ "description": "CoreML Model Weights",
13
+ "name": "weights",
14
+ "path": "com.apple.CoreML/weights"
15
+ }
16
+ },
17
+ "rootModelIdentifier": "912E1732-7E7E-40A7-BF38-EC7B6EB3B110"
18
+ }
vocab.json ADDED
@@ -0,0 +1,258 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ "<unk>",
3
+ "▁e",
4
+ "▁d",
5
+ "▁de",
6
+ "▁l",
7
+ "▁c",
8
+ "▁s",
9
+ "▁p",
10
+ "os",
11
+ "en",
12
+ "▁a",
13
+ "er",
14
+ "on",
15
+ "es",
16
+ "ra",
17
+ "▁la",
18
+ "ci",
19
+ "ta",
20
+ "or",
21
+ "▁es",
22
+ "▁en",
23
+ "▁m",
24
+ "ar",
25
+ "in",
26
+ "te",
27
+ "as",
28
+ "ad",
29
+ "un",
30
+ "an",
31
+ "▁el",
32
+ "re",
33
+ "al",
34
+ "ue",
35
+ "ti",
36
+ "▁f",
37
+ "▁se",
38
+ "▁y",
39
+ "▁un",
40
+ "is",
41
+ "▁su",
42
+ "▁con",
43
+ "▁h",
44
+ "to",
45
+ "▁n",
46
+ "ro",
47
+ "ón",
48
+ "do",
49
+ "ic",
50
+ "om",
51
+ "▁v",
52
+ "▁t",
53
+ "ri",
54
+ "id",
55
+ "▁b",
56
+ "▁re",
57
+ "ente",
58
+ "▁g",
59
+ "ción",
60
+ "ol",
61
+ "▁los",
62
+ "▁in",
63
+ "tu",
64
+ "▁al",
65
+ "▁com",
66
+ "que",
67
+ "▁del",
68
+ "▁o",
69
+ "la",
70
+ "res",
71
+ "di",
72
+ "le",
73
+ "▁por",
74
+ "mente",
75
+ "de",
76
+ "cu",
77
+ "ent",
78
+ "mp",
79
+ "▁una",
80
+ "ía",
81
+ "▁que",
82
+ "▁las",
83
+ "ado",
84
+ "ter",
85
+ "▁pa",
86
+ "ia",
87
+ "il",
88
+ "ran",
89
+ "▁pro",
90
+ "cion",
91
+ "▁fue",
92
+ "ma",
93
+ "▁ha",
94
+ "▁ca",
95
+ "ig",
96
+ "▁esta",
97
+ "da",
98
+ "▁j",
99
+ "ec",
100
+ "lo",
101
+ "us",
102
+ "ás",
103
+ "▁no",
104
+ "io",
105
+ "▁cu",
106
+ "ca",
107
+ "ica",
108
+ "▁des",
109
+ "ada",
110
+ "▁par",
111
+ "▁per",
112
+ "▁ac",
113
+ "iv",
114
+ "tra",
115
+ "mb",
116
+ "qu",
117
+ "▁como",
118
+ "cia",
119
+ "▁est",
120
+ "el",
121
+ "▁para",
122
+ "era",
123
+ "▁mu",
124
+ "iz",
125
+ "ba",
126
+ "ch",
127
+ "enta",
128
+ "tor",
129
+ "go",
130
+ "▁ti",
131
+ "idad",
132
+ "co",
133
+ "ita",
134
+ "ru",
135
+ "ien",
136
+ "▁ex",
137
+ "pa",
138
+ "tiv",
139
+ "▁sus",
140
+ "mo",
141
+ "▁tra",
142
+ "▁to",
143
+ "dad",
144
+ "▁an",
145
+ "ciones",
146
+ "▁ar",
147
+ "tos",
148
+ "▁di",
149
+ "▁cons",
150
+ "▁si",
151
+ "ales",
152
+ "▁ma",
153
+ "ul",
154
+ "▁más",
155
+ "án",
156
+ "almente",
157
+ "tan",
158
+ "dos",
159
+ "▁me",
160
+ "ier",
161
+ "▁son",
162
+ "▁lo",
163
+ "ici",
164
+ "▁te",
165
+ "uer",
166
+ "ce",
167
+ "▁dis",
168
+ "▁na",
169
+ "bre",
170
+ "ir",
171
+ "ión",
172
+ "rim",
173
+ "tro",
174
+ "▁as",
175
+ "▁ser",
176
+ "tes",
177
+ "man",
178
+ "▁po",
179
+ "les",
180
+ "ten",
181
+ "tas",
182
+ "ie",
183
+ "gu",
184
+ "ers",
185
+ "tura",
186
+ "▁este",
187
+ "ros",
188
+ "▁or",
189
+ "▁le",
190
+ "lan",
191
+ "du",
192
+ "▁col",
193
+ "ari",
194
+ "pe",
195
+ "ene",
196
+ "po",
197
+ "▁pr",
198
+ "ina",
199
+ "jo",
200
+ "▁prim",
201
+ "ores",
202
+ "▁so",
203
+ "▁res",
204
+ "im",
205
+ "▁ent",
206
+ "▁ob",
207
+ "ico",
208
+ "▁está",
209
+ "gun",
210
+ "▁car",
211
+ "za",
212
+ "ras",
213
+ "ido",
214
+ "▁dos",
215
+ "▁i",
216
+ "vo",
217
+ "ble",
218
+ "ga",
219
+ "ida",
220
+ "▁ta",
221
+ "ados",
222
+ "▁for",
223
+ "am",
224
+ "▁",
225
+ "e",
226
+ "a",
227
+ "o",
228
+ "s",
229
+ "n",
230
+ "r",
231
+ "i",
232
+ "l",
233
+ "d",
234
+ "t",
235
+ "c",
236
+ "u",
237
+ "m",
238
+ "p",
239
+ "b",
240
+ "g",
241
+ "v",
242
+ "f",
243
+ "ó",
244
+ "y",
245
+ "h",
246
+ "í",
247
+ "á",
248
+ "j",
249
+ "z",
250
+ "q",
251
+ "ñ",
252
+ "x",
253
+ "ú",
254
+ "k",
255
+ "w",
256
+ "é",
257
+ "ü"
258
+ ]