system HF staff commited on
Commit
c68e203
0 Parent(s):

Update files from the datasets library (from 1.17.0)

Browse files

Release notes: https://github.com/huggingface/datasets/releases/tag/1.17.0

.gitattributes ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bin.* filter=lfs diff=lfs merge=lfs -text
5
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.model filter=lfs diff=lfs merge=lfs -text
12
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
13
+ *.onnx filter=lfs diff=lfs merge=lfs -text
14
+ *.ot filter=lfs diff=lfs merge=lfs -text
15
+ *.parquet filter=lfs diff=lfs merge=lfs -text
16
+ *.pb filter=lfs diff=lfs merge=lfs -text
17
+ *.pt filter=lfs diff=lfs merge=lfs -text
18
+ *.pth filter=lfs diff=lfs merge=lfs -text
19
+ *.rar filter=lfs diff=lfs merge=lfs -text
20
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
21
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
22
+ *.tflite filter=lfs diff=lfs merge=lfs -text
23
+ *.tgz filter=lfs diff=lfs merge=lfs -text
24
+ *.xz filter=lfs diff=lfs merge=lfs -text
25
+ *.zip filter=lfs diff=lfs merge=lfs -text
26
+ *.zstandard filter=lfs diff=lfs merge=lfs -text
27
+ *tfevents* filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,282 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ annotations_creators:
3
+ - other
4
+ language_creators:
5
+ - crowdsourced
6
+ languages:
7
+ - en
8
+ licenses:
9
+ - cc-by-4-0
10
+ multilinguality:
11
+ - monolingual
12
+ pretty_name: SpeechCommands
13
+ size_categories:
14
+ v0-01:
15
+ - 10K<n<100K
16
+ v0-02:
17
+ - 100K<n<1M
18
+ source_datasets:
19
+ - original
20
+ task_categories:
21
+ - speech-processing
22
+ task_ids:
23
+ - other-other-keyword-spotting
24
+ ---
25
+
26
+ # Dataset Card for SpeechCommands
27
+
28
+ ## Table of Contents
29
+ - [Table of Contents](#table-of-contents)
30
+ - [Dataset Description](#dataset-description)
31
+ - [Dataset Summary](#dataset-summary)
32
+ - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
33
+ - [Languages](#languages)
34
+ - [Dataset Structure](#dataset-structure)
35
+ - [Data Instances](#data-instances)
36
+ - [Data Fields](#data-fields)
37
+ - [Data Splits](#data-splits)
38
+ - [Dataset Creation](#dataset-creation)
39
+ - [Curation Rationale](#curation-rationale)
40
+ - [Source Data](#source-data)
41
+ - [Annotations](#annotations)
42
+ - [Personal and Sensitive Information](#personal-and-sensitive-information)
43
+ - [Considerations for Using the Data](#considerations-for-using-the-data)
44
+ - [Social Impact of Dataset](#social-impact-of-dataset)
45
+ - [Discussion of Biases](#discussion-of-biases)
46
+ - [Other Known Limitations](#other-known-limitations)
47
+ - [Additional Information](#additional-information)
48
+ - [Dataset Curators](#dataset-curators)
49
+ - [Licensing Information](#licensing-information)
50
+ - [Citation Information](#citation-information)
51
+ - [Contributions](#contributions)
52
+
53
+ ## Dataset Description
54
+
55
+ - **Homepage:** https://www.tensorflow.org/datasets/catalog/speech_commands
56
+ - **Repository:** [More Information Needed]
57
+ - **Paper:** [Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition](https://arxiv.org/pdf/1804.03209.pdf)
58
+ - **Leaderboard:** [More Information Needed]
59
+ - **Point of Contact:** Pete Warden, petewarden@google.com
60
+
61
+ ### Dataset Summary
62
+
63
+ This is a set of one-second .wav audio files, each containing a single spoken
64
+ English word or background noise. These words are from a small set of commands, and are spoken by a
65
+ variety of different speakers. This data set is designed to help train simple
66
+ machine learning models. It is covered in more detail at [https://arxiv.org/abs/1804.03209](https://arxiv.org/abs/1804.03209).
67
+
68
+ Version 0.01 of the data set (configuration `"v0.01"`) was released on August 3rd 2017 and contains
69
+ 64,727 audio files.
70
+
71
+ Version 0.02 of the data set (configuration `"v0.02"`) was released on April 11th 2018 and
72
+ contains 105,829 audio files.
73
+
74
+
75
+ ### Supported Tasks and Leaderboards
76
+
77
+ * `keyword-spotting`: the dataset can be used to train and evaluate keyword
78
+ spotting systems. The task is to detect preregistered keywords by classifying utterances
79
+ into a predefined set of words. The task is usually performed on-device for the
80
+ fast response time. Thus, accuracy, model size, and inference time are all crucial.
81
+
82
+ ### Languages
83
+
84
+ The language data in SpeechCommands is in English (BCP-47 `en`).
85
+
86
+ ## Dataset Structure
87
+
88
+ ### Data Instances
89
+
90
+ Example of a core word (`"label"` is a word, `"is_unknown"` is `False`):
91
+ ```python
92
+ {
93
+ "file": "no/7846fd85_nohash_0.wav",
94
+ "audio": {
95
+ "path": "no/7846fd85_nohash_0.wav",
96
+ "array": array([ -0.00021362, -0.00027466, -0.00036621, ..., 0.00079346,
97
+ 0.00091553, 0.00079346]),
98
+ "sampling_rate": 16000
99
+ },
100
+ "label": 1, # "no"
101
+ "is_unknown": False,
102
+ "speaker_id": "7846fd85",
103
+ "utterance_id": 0
104
+ }
105
+ ```
106
+
107
+ Example of an auxiliary word (`"label"` is a word, `"is_unknown"` is `True`)
108
+ ```python
109
+ {
110
+ "file": "tree/8b775397_nohash_0.wav",
111
+ "audio": {
112
+ "path": "tree/8b775397_nohash_0.wav",
113
+ "array": array([ -0.00854492, -0.01339722, -0.02026367, ..., 0.00274658,
114
+ 0.00335693, 0.0005188]),
115
+ "sampling_rate": 16000
116
+ },
117
+ "label": 28, # "tree"
118
+ "is_unknown": True,
119
+ "speaker_id": "1b88bf70",
120
+ "utterance_id": 0
121
+ }
122
+ ```
123
+
124
+ Example of background noise (`_silence_`) class:
125
+
126
+ ```python
127
+ {
128
+ "file": "_silence_/doing_the_dishes.wav",
129
+ "audio": {
130
+ "path": "_silence_/doing_the_dishes.wav",
131
+ "array": array([ 0. , 0. , 0. , ..., -0.00592041,
132
+ -0.00405884, -0.00253296]),
133
+ "sampling_rate": 16000
134
+ },
135
+ "label": 30, # "_silence_"
136
+ "is_unknown": False,
137
+ "speaker_id": "None",
138
+ "utterance_id": 0 # doesn't make sense here
139
+ }
140
+ ```
141
+
142
+ ### Data Fields
143
+
144
+ * `file`: relative audio filename inside the original archive.
145
+ * `audio`: dictionary containing a relative audio filename,
146
+ a decoded audio array, and the sampling rate. Note that when accessing
147
+ the audio column: `dataset[0]["audio"]` the audio is automatically decoded
148
+ and resampled to `dataset.features["audio"].sampling_rate`.
149
+ Decoding and resampling of a large number of audios might take a significant
150
+ amount of time. Thus, it is important to first query the sample index before
151
+ the `"audio"` column, i.e. `dataset[0]["audio"]` should always be preferred
152
+ over `dataset["audio"][0]`.
153
+ * `label`: either word pronounced in an audio sample or background noise (`_silence_`) class.
154
+ Note that it's an integer value corresponding to the class name.
155
+ * `is_unknown`: if a word is auxiliary. Equals to `False` if a word is a core word or `_silence_`,
156
+ `True` if a word is an auxiliary word.
157
+ * `speaker_id`: unique id of a speaker. Equals to `None` if label is `_silence_`.
158
+ * `utterance_id`: incremental id of a word utterance within the same speaker.
159
+
160
+ ### Data Splits
161
+
162
+ The dataset has two versions (= configurations): `"v0.01"` and `"v0.02"`. `"v0.02"`
163
+ contains more words (see section [Source Data](#source-data) for more details).
164
+
165
+ | | train | validation | test |
166
+ |----- |------:|-----------:|-----:|
167
+ | v0.01 | 51093 | 6799 | 3081 |
168
+ | v0.02 | 84848 | 9982 | 4890 |
169
+
170
+ Note that in train and validation sets examples of `_silence_` class are longer than 1 second.
171
+ You can use the following code to sample 1-second examples from the longer ones:
172
+
173
+ ```python
174
+ def sample_noise(example):
175
+ # Use this function to extract random 1 sec slices of each _silence_ utterance,
176
+ # e.g. inside `torch.utils.data.Dataset.__getitem__()`
177
+ from random import randint
178
+
179
+ if example["label"] == "_silence_":
180
+ random_offset = randint(0, len(example["speech"]) - example["sample_rate"] - 1)
181
+ example["speech"] = example["speech"][random_offset : random_offset + example["sample_rate"]]
182
+
183
+ return example
184
+ ```
185
+
186
+ ## Dataset Creation
187
+
188
+ ### Curation Rationale
189
+
190
+ The primary goal of the dataset is to provide a way to build and test small
191
+ models that can detect a single word from a set of target words and differentiate it
192
+ from background noise or unrelated speech with as few false positives as possible.
193
+
194
+ ### Source Data
195
+
196
+ #### Initial Data Collection and Normalization
197
+
198
+ The audio files were collected using crowdsourcing, see
199
+ [aiyprojects.withgoogle.com/open_speech_recording](https://github.com/petewarden/extract_loudest_section)
200
+ for some of the open source audio collection code that was used. The goal was to gather examples of
201
+ people speaking single-word commands, rather than conversational sentences, so
202
+ they were prompted for individual words over the course of a five minute
203
+ session.
204
+
205
+ In version 0.01 thirty different words were recoded: "Yes", "No", "Up", "Down", "Left",
206
+ "Right", "On", "Off", "Stop", "Go", "Zero", "One", "Two", "Three", "Four", "Five", "Six", "Seven", "Eight", "Nine",
207
+ "Bed", "Bird", "Cat", "Dog", "Happy", "House", "Marvin", "Sheila", "Tree", "Wow".
208
+
209
+
210
+ In version 0.02 more words were added: "Backward", "Forward", "Follow", "Learn", "Visual".
211
+
212
+ In both versions, ten of them are used as commands by convention: "Yes", "No", "Up", "Down", "Left",
213
+ "Right", "On", "Off", "Stop", "Go". Other words are considered to be auxiliary (in current implementation
214
+ it is marked by `True` value of `"is_unknown"` feature). Their function is to teach a model to distinguish core words
215
+ from unrecognized ones.
216
+
217
+ The `_silence_` label contains a set of longer audio clips that are either recordings or
218
+ a mathematical simulation of noise.
219
+
220
+ #### Who are the source language producers?
221
+
222
+ The audio files were collected using crowdsourcing.
223
+
224
+ ### Annotations
225
+
226
+ #### Annotation process
227
+
228
+ Labels are the list of words prepared in advances.
229
+ Speakers were prompted for individual words over the course of a five minute
230
+ session.
231
+
232
+ #### Who are the annotators?
233
+
234
+ [More Information Needed]
235
+
236
+ ### Personal and Sensitive Information
237
+
238
+ [More Information Needed]
239
+
240
+ ## Considerations for Using the Data
241
+
242
+ ### Social Impact of Dataset
243
+
244
+ [More Information Needed]
245
+
246
+ ### Discussion of Biases
247
+
248
+ [More Information Needed]
249
+
250
+ ### Other Known Limitations
251
+
252
+ [More Information Needed]
253
+
254
+ ## Additional Information
255
+
256
+ ### Dataset Curators
257
+
258
+ [More Information Needed]
259
+
260
+ ### Licensing Information
261
+
262
+ Creative Commons BY 4.0 License.
263
+
264
+ ### Citation Information
265
+
266
+ ```
267
+ @article{speechcommandsv2,
268
+ author = { {Warden}, P.},
269
+ title = "{Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition}",
270
+ journal = {ArXiv e-prints},
271
+ archivePrefix = "arXiv",
272
+ eprint = {1804.03209},
273
+ primaryClass = "cs.CL",
274
+ keywords = {Computer Science - Computation and Language, Computer Science - Human-Computer Interaction},
275
+ year = 2018,
276
+ month = apr,
277
+ url = {https://arxiv.org/abs/1804.03209},
278
+ }
279
+ ```
280
+ ### Contributions
281
+
282
+ Thanks to [@polinaeterna](https://github.com/polinaeterna) for adding this dataset.
dataset_infos.json ADDED
@@ -0,0 +1 @@
 
1
+ {"v0.01": {"description": "\nThis is a set of one-second .wav audio files, each containing a single spoken\nEnglish word or background noise. These words are from a small set of commands, and are spoken by a\nvariety of different speakers. This data set is designed to help train simple\nmachine learning models. This dataset is covered in more detail at\n[https://arxiv.org/abs/1804.03209](https://arxiv.org/abs/1804.03209).\n\nVersion 0.01 of the data set (configuration `\"v0.01\"`) was released on August 3rd 2017 and contains\n64,727 audio files.\n\nIn version 0.01 thirty different words were recoded: \"Yes\", \"No\", \"Up\", \"Down\", \"Left\",\n\"Right\", \"On\", \"Off\", \"Stop\", \"Go\", \"Zero\", \"One\", \"Two\", \"Three\", \"Four\", \"Five\", \"Six\", \"Seven\", \"Eight\", \"Nine\",\n\"Bed\", \"Bird\", \"Cat\", \"Dog\", \"Happy\", \"House\", \"Marvin\", \"Sheila\", \"Tree\", \"Wow\".\n\n\nIn version 0.02 more words were added: \"Backward\", \"Forward\", \"Follow\", \"Learn\", \"Visual\".\n\nIn both versions, ten of them are used as commands by convention: \"Yes\", \"No\", \"Up\", \"Down\", \"Left\",\n\"Right\", \"On\", \"Off\", \"Stop\", \"Go\". Other words are considered to be auxiliary (in current implementation\nit is marked by `True` value of `\"is_unknown\"` feature). Their function is to teach a model to distinguish core words\nfrom unrecognized ones.\n\nThe `_silence_` class contains a set of longer audio clips that are either recordings or\na mathematical simulation of noise.\n\n", "citation": "\n@article{speechcommandsv2,\n author = { {Warden}, P.},\n title = \"{Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition}\",\n journal = {ArXiv e-prints},\n archivePrefix = \"arXiv\",\n eprint = {1804.03209},\n primaryClass = \"cs.CL\",\n keywords = {Computer Science - Computation and Language, Computer Science - Human-Computer Interaction},\n year = 2018,\n month = apr,\n url = {https://arxiv.org/abs/1804.03209},\n}\n", "homepage": "https://www.tensorflow.org/datasets/catalog/speech_commands", "license": "Creative Commons BY 4.0 License", "features": {"file": {"dtype": "string", "id": null, "_type": "Value"}, "audio": {"sampling_rate": 16000, "mono": true, "_storage_dtype": "struct", "id": null, "_type": "Audio"}, "label": {"num_classes": 31, "names": ["yes", "no", "up", "down", "left", "right", "on", "off", "stop", "go", "zero", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine", "bed", "bird", "cat", "dog", "happy", "house", "marvin", "sheila", "tree", "wow", "_silence_"], "names_file": null, "id": null, "_type": "ClassLabel"}, "is_unknown": {"dtype": "bool", "id": null, "_type": "Value"}, "speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "utterance_id": {"dtype": "int8", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "speech_commands", "config_name": "v0.01", "version": {"version_str": "0.1.0", "description": null, "major": 0, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 1626283624, "num_examples": 51093, "dataset_name": "speech_commands"}, "validation": {"name": "validation", "num_bytes": 217204539, "num_examples": 6799, "dataset_name": "speech_commands"}, "test": {"name": "test", "num_bytes": 98979965, "num_examples": 3081, "dataset_name": "speech_commands"}}, "download_checksums": {"https://s3.amazonaws.com/datasets.huggingface.co/SpeechCommands/v0.01/v0.01_train.tar.gz": {"num_bytes": 1176079446, "checksum": "41e4893c0750e576fae0b9c2bb066f6b93b8b7be7be58c60faf037835e7ccb3b"}, "https://s3.amazonaws.com/datasets.huggingface.co/SpeechCommands/v0.01/v0.01_validation.tar.gz": {"num_bytes": 153663590, "checksum": "8f9e1dd79699be3b34552960aa03af1e1c24b4ee014ca28d18d01faca72e1dc0"}, "https://s3.amazonaws.com/datasets.huggingface.co/SpeechCommands/v0.01/v0.01_test.tar.gz": {"num_bytes": 124959719, "checksum": "0ccb217235db6b9e12a3ba3436180bac56d488155ce857614d14584d2b6a3546"}}, "download_size": 1454702755, "post_processing_size": null, "dataset_size": 1942468128, "size_in_bytes": 3397170883}, "v0.02": {"description": "\nThis is a set of one-second .wav audio files, each containing a single spoken\nEnglish word or background noise. These words are from a small set of commands, and are spoken by a\nvariety of different speakers. This data set is designed to help train simple\nmachine learning models. This dataset is covered in more detail at\n[https://arxiv.org/abs/1804.03209](https://arxiv.org/abs/1804.03209).\n\nVersion 0.01 of the data set (configuration `\"v0.01\"`) was released on August 3rd 2017 and contains\n64,727 audio files.\n\nIn version 0.01 thirty different words were recoded: \"Yes\", \"No\", \"Up\", \"Down\", \"Left\",\n\"Right\", \"On\", \"Off\", \"Stop\", \"Go\", \"Zero\", \"One\", \"Two\", \"Three\", \"Four\", \"Five\", \"Six\", \"Seven\", \"Eight\", \"Nine\",\n\"Bed\", \"Bird\", \"Cat\", \"Dog\", \"Happy\", \"House\", \"Marvin\", \"Sheila\", \"Tree\", \"Wow\".\n\n\nIn version 0.02 more words were added: \"Backward\", \"Forward\", \"Follow\", \"Learn\", \"Visual\".\n\nIn both versions, ten of them are used as commands by convention: \"Yes\", \"No\", \"Up\", \"Down\", \"Left\",\n\"Right\", \"On\", \"Off\", \"Stop\", \"Go\". Other words are considered to be auxiliary (in current implementation\nit is marked by `True` value of `\"is_unknown\"` feature). Their function is to teach a model to distinguish core words\nfrom unrecognized ones.\n\nThe `_silence_` class contains a set of longer audio clips that are either recordings or\na mathematical simulation of noise.\n\n", "citation": "\n@article{speechcommandsv2,\n author = { {Warden}, P.},\n title = \"{Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition}\",\n journal = {ArXiv e-prints},\n archivePrefix = \"arXiv\",\n eprint = {1804.03209},\n primaryClass = \"cs.CL\",\n keywords = {Computer Science - Computation and Language, Computer Science - Human-Computer Interaction},\n year = 2018,\n month = apr,\n url = {https://arxiv.org/abs/1804.03209},\n}\n", "homepage": "https://www.tensorflow.org/datasets/catalog/speech_commands", "license": "Creative Commons BY 4.0 License", "features": {"file": {"dtype": "string", "id": null, "_type": "Value"}, "audio": {"sampling_rate": 16000, "mono": true, "_storage_dtype": "struct", "id": null, "_type": "Audio"}, "label": {"num_classes": 36, "names": ["yes", "no", "up", "down", "left", "right", "on", "off", "stop", "go", "zero", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine", "bed", "bird", "cat", "dog", "happy", "house", "marvin", "sheila", "tree", "wow", "backward", "forward", "follow", "learn", "visual", "_silence_"], "names_file": null, "id": null, "_type": "ClassLabel"}, "is_unknown": {"dtype": "bool", "id": null, "_type": "Value"}, "speaker_id": {"dtype": "string", "id": null, "_type": "Value"}, "utterance_id": {"dtype": "int8", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "speech_commands", "config_name": "v0.02", "version": {"version_str": "0.2.0", "description": null, "major": 0, "minor": 2, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2684381672, "num_examples": 84848, "dataset_name": "speech_commands"}, "validation": {"name": "validation", "num_bytes": 316435178, "num_examples": 9982, "dataset_name": "speech_commands"}, "test": {"name": "test", "num_bytes": 157096106, "num_examples": 4890, "dataset_name": "speech_commands"}}, "download_checksums": {"https://s3.amazonaws.com/datasets.huggingface.co/SpeechCommands/v0.02/v0.02_train.tar.gz": {"num_bytes": 1944462432, "checksum": "acfc1a9e5f020ef5d20f13bb5c1035dcc19a3cc6d5fd1fe775d99814ce840399"}, "https://s3.amazonaws.com/datasets.huggingface.co/SpeechCommands/v0.02/v0.02_validation.tar.gz": {"num_bytes": 229117586, "checksum": "868bdecd3dc12276ee55d2aeca5b1f02d913d6f17875181c1bf9d465fa2f7be1"}, "https://s3.amazonaws.com/datasets.huggingface.co/SpeechCommands/v0.02/v0.02_test.tar.gz": {"num_bytes": 112395851, "checksum": "45aedb39cb2c9f03e098a8d5c98350d6d8473c432ad4558fce26c6feb478a812"}}, "download_size": 2285975869, "post_processing_size": null, "dataset_size": 3157912956, "size_in_bytes": 5443888825}}
dummy/v0.01/0.1.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:39ca9ffcaa077c432fc12db15f532bd64763809cbe96743e8719647cba993137
3
+ size 3159718
dummy/v0.02/0.2.0/dummy_data.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb7ec694a31b76eddaf8a026a2ff077a5e50e17aeb51f678aeaf170067f294f9
3
+ size 3346327
speech_commands.py ADDED
@@ -0,0 +1,229 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # coding=utf-8
2
+ # Copyright 2021 The HuggingFace Datasets Authors and the current dataset script contributor.
3
+ #
4
+ # Licensed under the Apache License, Version 2.0 (the "License");
5
+ # you may not use this file except in compliance with the License.
6
+ # You may obtain a copy of the License at
7
+ #
8
+ # http://www.apache.org/licenses/LICENSE-2.0
9
+ #
10
+ # Unless required by applicable law or agreed to in writing, software
11
+ # distributed under the License is distributed on an "AS IS" BASIS,
12
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
+ # See the License for the specific language governing permissions and
14
+ # limitations under the License.
15
+
16
+ """Speech Commands, an audio dataset of spoken words designed to help train and evaluate keyword spotting systems. """
17
+
18
+
19
+ import textwrap
20
+
21
+ import datasets
22
+
23
+
24
+ _CITATION = """
25
+ @article{speechcommandsv2,
26
+ author = { {Warden}, P.},
27
+ title = "{Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition}",
28
+ journal = {ArXiv e-prints},
29
+ archivePrefix = "arXiv",
30
+ eprint = {1804.03209},
31
+ primaryClass = "cs.CL",
32
+ keywords = {Computer Science - Computation and Language, Computer Science - Human-Computer Interaction},
33
+ year = 2018,
34
+ month = apr,
35
+ url = {https://arxiv.org/abs/1804.03209},
36
+ }
37
+ """
38
+
39
+ _DESCRIPTION = """
40
+ This is a set of one-second .wav audio files, each containing a single spoken
41
+ English word or background noise. These words are from a small set of commands, and are spoken by a
42
+ variety of different speakers. This data set is designed to help train simple
43
+ machine learning models. This dataset is covered in more detail at
44
+ [https://arxiv.org/abs/1804.03209](https://arxiv.org/abs/1804.03209).
45
+
46
+ Version 0.01 of the data set (configuration `"v0.01"`) was released on August 3rd 2017 and contains
47
+ 64,727 audio files.
48
+
49
+ In version 0.01 thirty different words were recoded: "Yes", "No", "Up", "Down", "Left",
50
+ "Right", "On", "Off", "Stop", "Go", "Zero", "One", "Two", "Three", "Four", "Five", "Six", "Seven", "Eight", "Nine",
51
+ "Bed", "Bird", "Cat", "Dog", "Happy", "House", "Marvin", "Sheila", "Tree", "Wow".
52
+
53
+
54
+ In version 0.02 more words were added: "Backward", "Forward", "Follow", "Learn", "Visual".
55
+
56
+ In both versions, ten of them are used as commands by convention: "Yes", "No", "Up", "Down", "Left",
57
+ "Right", "On", "Off", "Stop", "Go". Other words are considered to be auxiliary (in current implementation
58
+ it is marked by `True` value of `"is_unknown"` feature). Their function is to teach a model to distinguish core words
59
+ from unrecognized ones.
60
+
61
+ The `_silence_` class contains a set of longer audio clips that are either recordings or
62
+ a mathematical simulation of noise.
63
+
64
+ """
65
+
66
+ _LICENSE = "Creative Commons BY 4.0 License"
67
+
68
+ _URL = "https://www.tensorflow.org/datasets/catalog/speech_commands"
69
+
70
+ _DL_URL = "https://s3.amazonaws.com/datasets.huggingface.co/SpeechCommands/{name}/{name}_{split}.tar.gz"
71
+
72
+ WORDS = [
73
+ "yes",
74
+ "no",
75
+ "up",
76
+ "down",
77
+ "left",
78
+ "right",
79
+ "on",
80
+ "off",
81
+ "stop",
82
+ "go",
83
+ ]
84
+
85
+ UNKNOWN_WORDS_V1 = [
86
+ "zero",
87
+ "one",
88
+ "two",
89
+ "three",
90
+ "four",
91
+ "five",
92
+ "six",
93
+ "seven",
94
+ "eight",
95
+ "nine",
96
+ "bed",
97
+ "bird",
98
+ "cat",
99
+ "dog",
100
+ "happy",
101
+ "house",
102
+ "marvin",
103
+ "sheila",
104
+ "tree",
105
+ "wow",
106
+ ]
107
+
108
+ UNKNOWN_WORDS_V2 = UNKNOWN_WORDS_V1 + [
109
+ "backward",
110
+ "forward",
111
+ "follow",
112
+ "learn",
113
+ "visual",
114
+ ]
115
+
116
+ SILENCE = "_silence_" # background noise
117
+ LABELS_V1 = WORDS + UNKNOWN_WORDS_V1 + [SILENCE]
118
+ LABELS_V2 = WORDS + UNKNOWN_WORDS_V2 + [SILENCE]
119
+
120
+
121
+ class SpeechCommandsConfig(datasets.BuilderConfig):
122
+ """BuilderConfig for SpeechCommands."""
123
+
124
+ def __init__(self, labels, **kwargs):
125
+ super(SpeechCommandsConfig, self).__init__(**kwargs)
126
+ self.labels = labels
127
+
128
+
129
+ class SpeechCommands(datasets.GeneratorBasedBuilder):
130
+ BUILDER_CONFIGS = [
131
+ SpeechCommandsConfig(
132
+ name="v0.01",
133
+ description=textwrap.dedent(
134
+ """\
135
+ Version 0.01 of the SpeechCommands dataset. Contains 30 words
136
+ (20 of them are auxiliary) and background noise.
137
+ """
138
+ ),
139
+ labels=LABELS_V1,
140
+ version=datasets.Version("0.1.0"),
141
+ ),
142
+ SpeechCommandsConfig(
143
+ name="v0.02",
144
+ description=textwrap.dedent(
145
+ """\
146
+ Version 0.02 of the SpeechCommands dataset.
147
+ Contains 35 words (25 of them are auxiliary) and background noise.
148
+ """
149
+ ),
150
+ labels=LABELS_V2,
151
+ version=datasets.Version("0.2.0"),
152
+ ),
153
+ ]
154
+
155
+ def _info(self):
156
+ return datasets.DatasetInfo(
157
+ description=_DESCRIPTION,
158
+ features=datasets.Features(
159
+ {
160
+ "file": datasets.Value("string"),
161
+ "audio": datasets.features.Audio(sampling_rate=16_000),
162
+ "label": datasets.ClassLabel(names=self.config.labels),
163
+ "is_unknown": datasets.Value("bool"),
164
+ "speaker_id": datasets.Value("string"),
165
+ "utterance_id": datasets.Value("int8"),
166
+ }
167
+ ),
168
+ homepage=_URL,
169
+ citation=_CITATION,
170
+ license=_LICENSE,
171
+ version=self.config.version,
172
+ )
173
+
174
+ def _split_generators(self, dl_manager):
175
+
176
+ archive_paths = dl_manager.download(
177
+ {
178
+ "train": _DL_URL.format(name=self.config.name, split="train"),
179
+ "validation": _DL_URL.format(name=self.config.name, split="validation"),
180
+ "test": _DL_URL.format(name=self.config.name, split="test"),
181
+ }
182
+ )
183
+
184
+ return [
185
+ datasets.SplitGenerator(
186
+ name=datasets.Split.TRAIN,
187
+ gen_kwargs={
188
+ "archive": dl_manager.iter_archive(archive_paths["train"]),
189
+ },
190
+ ),
191
+ datasets.SplitGenerator(
192
+ name=datasets.Split.VALIDATION,
193
+ gen_kwargs={
194
+ "archive": dl_manager.iter_archive(archive_paths["validation"]),
195
+ },
196
+ ),
197
+ datasets.SplitGenerator(
198
+ name=datasets.Split.TEST,
199
+ gen_kwargs={
200
+ "archive": dl_manager.iter_archive(archive_paths["test"]),
201
+ },
202
+ ),
203
+ ]
204
+
205
+ def _generate_examples(self, archive):
206
+ for path, file in archive:
207
+ if not path.endswith(".wav"):
208
+ continue
209
+
210
+ word, audio_filename = path.split("/")
211
+ is_unknown = False
212
+
213
+ if word == SILENCE:
214
+ speaker_id, utterance_id = None, 0
215
+
216
+ else: # word is either in WORDS or unknown
217
+ if word not in WORDS:
218
+ is_unknown = True
219
+ # an audio filename looks like `0bac8a71_nohash_0.wav`
220
+ speaker_id, _, utterance_id = audio_filename.split(".wav")[0].split("_")
221
+
222
+ yield path, {
223
+ "file": path,
224
+ "audio": {"path": path, "bytes": file.read()},
225
+ "label": word,
226
+ "is_unknown": is_unknown,
227
+ "speaker_id": speaker_id,
228
+ "utterance_id": utterance_id,
229
+ }