holylovenia commited on
Commit
7a44f43
1 Parent(s): df81181

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +41 -69
README.md CHANGED
@@ -30,21 +30,10 @@ task_ids:
30
  - [Dataset Summary](#dataset-summary)
31
  - [Supported Tasks](#supported-tasks-and-leaderboards)
32
  - [Languages](#languages)
 
33
  - [Dataset Structure](#dataset-structure)
34
- - [Data Instances](#data-instances)
35
- - [Data Fields](#data-instances)
36
  - [Data Splits](#data-instances)
37
- - [Dataset Creation](#dataset-creation)
38
- - [Curation Rationale](#curation-rationale)
39
- - [Source Data](#source-data)
40
- - [Annotations](#annotations)
41
- - [Personal and Sensitive Information](#personal-and-sensitive-information)
42
- - [Considerations for Using the Data](#considerations-for-using-the-data)
43
- - [Social Impact of Dataset](#social-impact-of-dataset)
44
- - [Discussion of Biases](#discussion-of-biases)
45
- - [Other Known Limitations](#other-known-limitations)
46
  - [Additional Information](#additional-information)
47
- - [Dataset Curators](#dataset-curators)
48
  - [Licensing Information](#licensing-information)
49
  - [Citation Information](#citation-information)
50
 
@@ -62,80 +51,63 @@ ASCEND (A Spontaneous Chinese-English Dataset) introduces a high-quality resourc
62
 
63
  ### Supported Tasks and Leaderboards
64
 
65
- [Needs More Information]
66
 
67
  ### Languages
68
 
69
  Chinese and English
70
 
71
- ## Dataset Structure
72
-
73
- ### Data Instances
74
 
75
- [Needs More Information]
 
 
 
76
 
77
- ### Data Fields
78
 
79
- [Needs More Information]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
80
 
81
  ### Data Splits
82
 
83
- [Needs More Information]
84
-
85
- ## Dataset Creation
86
-
87
- ### Curation Rationale
88
-
89
- [Needs More Information]
90
-
91
- ### Source Data
92
-
93
- #### Initial Data Collection and Normalization
94
-
95
- [Needs More Information]
96
-
97
- #### Who are the source language producers?
98
-
99
- [Needs More Information]
100
-
101
- ### Annotations
102
-
103
- #### Annotation process
104
-
105
- [Needs More Information]
106
-
107
- #### Who are the annotators?
108
-
109
- [Needs More Information]
110
-
111
- ### Personal and Sensitive Information
112
-
113
- [Needs More Information]
114
-
115
- ## Considerations for Using the Data
116
-
117
- ### Social Impact of Dataset
118
-
119
- [Needs More Information]
120
-
121
- ### Discussion of Biases
122
-
123
- [Needs More Information]
124
-
125
- ### Other Known Limitations
126
-
127
- [Needs More Information]
128
 
129
  ## Additional Information
130
 
131
- ### Dataset Curators
132
-
133
- [Needs More Information]
134
 
135
  ### Licensing Information
136
 
137
- [Needs More Information]
138
 
139
  ### Citation Information
140
 
141
- [Needs More Information]
 
 
 
 
 
 
 
 
 
30
  - [Dataset Summary](#dataset-summary)
31
  - [Supported Tasks](#supported-tasks-and-leaderboards)
32
  - [Languages](#languages)
33
+ - [Usage](#usage)
34
  - [Dataset Structure](#dataset-structure)
 
 
35
  - [Data Splits](#data-instances)
 
 
 
 
 
 
 
 
 
36
  - [Additional Information](#additional-information)
 
37
  - [Licensing Information](#licensing-information)
38
  - [Citation Information](#citation-information)
39
 
 
51
 
52
  ### Supported Tasks and Leaderboards
53
 
54
+ Code-switching.
55
 
56
  ### Languages
57
 
58
  Chinese and English
59
 
60
+ ## Usage
 
 
61
 
62
+ ```
63
+ import datasets
64
+ dataset = datasets.load_dataset("CAiRE/ASCEND", "train") # split: "train", "validation", "test"
65
+ ```
66
 
67
+ ## Dataset Structure
68
 
69
+ A typical data point comprises the path to the audio file, the loaded audio array, and its transcription. Additional fields include datapoint id, duration, language, speaker id, session id, and topic.
70
+
71
+ ```
72
+ {
73
+ 'id': '00000',
74
+ 'path': '.cache/huggingface/datasets/downloads/extracted/f0b33b5266cd9452ee310eef3577cf7adb7f29aa54dbff74b9a8ee406a55d614/waves/ses1_spk17_L3818_9.3200_0.6400.wav',
75
+ 'audio': {
76
+ 'path': '.cache/huggingface/datasets/downloads/extracted/f0b33b5266cd9452ee310eef3577cf7adb7f29aa54dbff74b9a8ee406a55d614/waves/ses1_spk17_L3818_9.3200_0.6400.wav',
77
+ 'array': array([0.00057983, 0.00073242, 0.00125122, ..., 0.00204468, 0.00250244,
78
+ 0.00201416
79
+ ], dtype = float32),
80
+ 'sampling_rate': 16000
81
+ },
82
+ 'transcription': '好的',
83
+ 'duration': 0.6399999856948853,
84
+ 'language': 'zh',
85
+ 'original_speaker_id': 17,
86
+ 'session_id': 1,
87
+ 'topic': 'persona'
88
+ }
89
+ ```
90
 
91
  ### Data Splits
92
 
93
+ Number of utterances: 9,869 train, 1,129 validation, and 1,315 test.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
94
 
95
  ## Additional Information
96
 
97
+ For comprehensive explanations, please check [our paper](https://arxiv.org/pdf/2112.06223.pdf).
 
 
98
 
99
  ### Licensing Information
100
 
101
+ Creative Common Attribution Share-Alike 4.0 International (CC-BY-SA 4.0)
102
 
103
  ### Citation Information
104
 
105
+ If you use our dataset, please cite us:
106
+
107
+ ```
108
+ @inproceedings{lovenia2022ascend,
109
+ title={ASCEND: A Spontaneous Chinese-English Dataset for Code-switching in Multi-turn Conversation},
110
+ author={Lovenia, Holy and Cahyawijaya, Samuel and Winata, Genta Indra and Xu, Peng and Yan, Xu and Liu, Zihan and Frieske, Rita and Yu, Tiezheng and Dai, Wenliang and Barezi, Elham J and others},
111
+ booktitle={Proceedings of the 13th Language Resources and Evaluation Conference (LREC)},
112
+ year={2022}
113
+ ```