fsicoli commited on
Commit
e8727a3
1 Parent(s): 1213380

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +190 -0
README.md ADDED
@@ -0,0 +1,190 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc0-1.0
3
+ language:
4
+ - ab
5
+ - af
6
+ - am
7
+ - ar
8
+ - as
9
+ - ast
10
+ - az
11
+ - ba
12
+ - bas
13
+ - be
14
+ - bg
15
+ - bn
16
+ - br
17
+ - ca
18
+ - ckb
19
+ - cnh
20
+ - cs
21
+ - cv
22
+ - cy
23
+ - da
24
+ - de
25
+ - dv
26
+ - dyu
27
+ - el
28
+ - en
29
+ - eo
30
+ - es
31
+ - et
32
+ - eu
33
+ - fa
34
+ - fi
35
+ - fr
36
+ - gl
37
+ - gn
38
+ - ha
39
+ - he
40
+ - hi
41
+ - hsb
42
+ - hu
43
+ - ia
44
+ - id
45
+ - ig
46
+ - is
47
+ - it
48
+ - ja
49
+ - ka
50
+ - kab
51
+ - kk
52
+ - kmr
53
+ - ko
54
+ - ky
55
+ - lg
56
+ - lo
57
+ - lt
58
+ - lv
59
+ - mdf
60
+ - mhr
61
+ - mk
62
+ - ml
63
+ - mn
64
+ - mr
65
+ - mrj
66
+ - mt
67
+ - myv
68
+ - nl
69
+ - oc
70
+ - or
71
+ - pl
72
+ - ps
73
+ - pt
74
+ - quy
75
+ - ro
76
+ - ru
77
+ - rw
78
+ - sah
79
+ - sat
80
+ - sc
81
+ - sk
82
+ - skr
83
+ - sl
84
+ - sq
85
+ - sr
86
+ - sw
87
+ - ta
88
+ - th
89
+ - ti
90
+ - tig
91
+ - tk
92
+ - tok
93
+ - tr
94
+ - tt
95
+ - tw
96
+ - ug
97
+ - uk
98
+ - ur
99
+ - uz
100
+ - vi
101
+ - vot
102
+ - yue
103
+ - za
104
+ - zgh
105
+ - zh
106
+ - yo
107
+ task_categories:
108
+ - automatic-speech-recognition
109
+ pretty_name: Common Voice Corpus 17.0
110
+ size_categories:
111
+ - 100B<n<1T
112
+ tags:
113
+ - mozilla
114
+ - foundation
115
+ ---
116
+ # Dataset Card for Common Voice Corpus 17.0
117
+
118
+ <!-- Provide a quick summary of the dataset. -->
119
+
120
+ This dataset is an unofficial version of the Mozilla Common Voice Corpus 17. It was downloaded and converted from the project's website https://commonvoice.mozilla.org/.
121
+
122
+
123
+ ## Languages
124
+ ```
125
+ Abkhaz, Albanian, Amharic, Arabic, Armenian, Assamese, Asturian, Azerbaijani, Basaa, Bashkir, Basque, Belarusian, Bengali, Breton, Bulgarian, Cantonese, Catalan, Central Kurdish, Chinese (China), Chinese (Hong Kong), Chinese (Taiwan), Chuvash, Czech, Danish, Dhivehi, Dioula, Dutch, English, Erzya, Esperanto, Estonian, Finnish, French, Frisian, Galician, Georgian, German, Greek, Guarani, Hakha Chin, Hausa, Hill Mari, Hindi, Hungarian, Icelandic, Igbo, Indonesian, Interlingua, Irish, Italian, Japanese, Kabyle, Kazakh, Kinyarwanda, Korean, Kurmanji Kurdish, Kyrgyz, Lao, Latvian, Lithuanian, Luganda, Macedonian, Malayalam, Maltese, Marathi, Meadow Mari, Moksha, Mongolian, Nepali, Norwegian Nynorsk, Occitan, Odia, Pashto, Persian, Polish, Portuguese, Punjabi, Quechua Chanka, Romanian, Romansh Sursilvan, Romansh Vallader, Russian, Sakha, Santali (Ol Chiki), Saraiki, Sardinian, Serbian, Slovak, Slovenian, Sorbian, Upper, Spanish, Swahili, Swedish, Taiwanese (Minnan), Tamazight, Tamil, Tatar, Thai, Tigre, Tigrinya, Toki Pona, Turkish, Turkmen, Twi, Ukrainian, Urdu, Uyghur, Uzbek, Vietnamese, Votic, Welsh, Yoruba
126
+ ```
127
+
128
+ ## How to use
129
+ The datasets library allows you to load and pre-process your dataset in pure Python, at scale. The dataset can be downloaded and prepared in one call to your local drive by using the load_dataset function.
130
+
131
+ For example, to download the Portuguese config, simply specify the corresponding language config name (i.e., "pt" for Portuguese):
132
+ ```
133
+ from datasets import load_dataset
134
+
135
+ cv_17 = load_dataset("fsicoli/common_voice_17_0", "pt", split="train")
136
+ ```
137
+ Using the datasets library, you can also stream the dataset on-the-fly by adding a streaming=True argument to the load_dataset function call. Loading a dataset in streaming mode loads individual samples of the dataset at a time, rather than downloading the entire dataset to disk.
138
+
139
+ ```
140
+ from datasets import load_dataset
141
+
142
+ cv_17 = load_dataset("fsicoli/common_voice_17_0", "pt", split="train", streaming=True)
143
+
144
+ print(next(iter(cv_17)))
145
+ ```
146
+
147
+ Bonus: create a PyTorch dataloader directly with your own datasets (local/streamed).
148
+
149
+ ### Local
150
+ ```
151
+ from datasets import load_dataset
152
+ from torch.utils.data.sampler import BatchSampler, RandomSampler
153
+
154
+ cv_17 = load_dataset("fsicoli/common_voice_17_0", "pt", split="train")
155
+ batch_sampler = BatchSampler(RandomSampler(cv_17), batch_size=32, drop_last=False)
156
+ dataloader = DataLoader(cv_17, batch_sampler=batch_sampler)
157
+ ```
158
+
159
+ ### Streaming
160
+ ```
161
+ from datasets import load_dataset
162
+ from torch.utils.data import DataLoader
163
+
164
+ cv_17 = load_dataset("fsicoli/common_voice_17_0", "pt", split="train")
165
+ dataloader = DataLoader(cv_17, batch_size=32)
166
+ ```
167
+
168
+ To find out more about loading and preparing audio datasets, head over to hf.co/blog/audio-datasets.
169
+
170
+
171
+
172
+ ### Dataset Structure
173
+ Data Instances
174
+ A typical data point comprises the path to the audio file and its sentence. Additional fields include accent, age, client_id, up_votes, down_votes, gender, locale and segment.
175
+
176
+ ### Licensing Information
177
+ Public Domain, CC-0
178
+
179
+
180
+ ### Citation Information
181
+ ```
182
+ @inproceedings{commonvoice:2020,
183
+ author = {Ardila, R. and Branson, M. and Davis, K. and Henretty, M. and Kohler, M. and Meyer, J. and Morais, R. and Saunders, L. and Tyers, F. M. and Weber, G.},
184
+ title = {Common Voice: A Massively-Multilingual Speech Corpus},
185
+ booktitle = {Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020)},
186
+ pages = {4211--4215},
187
+ year = 2020
188
+ }
189
+ ```
190
+ ---