Datasets:

Modalities:
Audio
Text
Formats:
parquet
Languages:
Malayalam
DOI:
Libraries:
Datasets
Dask
License:
kavyamanohar commited on
Commit
8a40db2
·
1 Parent(s): 176eb44

Add dataset details

Browse files
Files changed (1) hide show
  1. README.md +39 -40
README.md CHANGED
@@ -71,97 +71,96 @@ TODO: Add YAML tags here. Copy-paste the tags obtained with the online tagging a
71
 
72
  ## Dataset Description
73
 
74
- - **Homepage: https://smc.org.in**
75
  - **Repository: https://gitlab.com/smc/msc-reviewed-speech**
76
- - **Paper: https://blog.smc.org.in/malayalam-speech-corpus/**
77
  - **Leaderboard:**
78
  - **Point of Contact: Kavya Manohar**
79
 
80
  ### Dataset Summary
81
 
82
- [More Information Needed]
83
 
84
- ### Supported Tasks and Leaderboards
85
 
86
- [More Information Needed]
 
 
 
 
 
 
 
 
87
 
 
 
 
 
88
  ### Languages
89
 
90
- [More Information Needed]
91
 
92
  ## Dataset Structure
93
 
 
 
 
 
 
 
 
 
 
94
  ### Data Instances
95
 
96
- [More Information Needed]
97
 
98
  ### Data Fields
99
 
100
- [More Information Needed]
101
 
102
  ### Data Splits
103
 
104
- [More Information Needed]
105
 
106
  ## Dataset Creation
107
 
108
- ### Curation Rationale
109
-
110
- [More Information Needed]
111
 
112
- ### Source Data
113
-
114
- #### Initial Data Collection and Normalization
115
-
116
- [More Information Needed]
117
 
118
- #### Who are the source language producers?
119
 
120
- [More Information Needed]
121
 
122
- ### Annotations
123
 
124
- #### Annotation process
125
 
126
- [More Information Needed]
127
 
128
- #### Who are the annotators?
129
 
130
- [More Information Needed]
131
 
132
  ### Personal and Sensitive Information
133
 
134
- [More Information Needed]
135
 
136
  ## Considerations for Using the Data
137
 
138
  ### Social Impact of Dataset
139
 
140
- [More Information Needed]
141
-
142
- ### Discussion of Biases
143
-
144
- [More Information Needed]
145
-
146
- ### Other Known Limitations
147
-
148
- [More Information Needed]
149
 
150
- ## Additional Information
151
 
152
  ### Dataset Curators
153
 
154
- [More Information Needed]
155
 
156
  ### Licensing Information
157
 
158
- [More Information Needed]
159
 
160
  ### Citation Information
161
 
162
- [More Information Needed]
163
 
164
  ### Contributions
165
 
166
- Thanks to [@github-username](https://github.com/<github-username>) for adding this dataset.
167
- [More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
 
71
 
72
  ## Dataset Description
73
 
74
+ - **Homepage: [SMC](https://smc.org.in)**
75
  - **Repository: https://gitlab.com/smc/msc-reviewed-speech**
76
+ - **Blog Post: https://blog.smc.org.in/malayalam-speech-corpus/**
77
  - **Leaderboard:**
78
  - **Point of Contact: Kavya Manohar**
79
 
80
  ### Dataset Summary
81
 
 
82
 
 
83
 
84
+ - 1541 speech samples
85
+ - 75 speech contributors
86
+ - 1:38:16 hours of speech
87
+ - 482 unique sentences
88
+ - 1400 unique words
89
+ - 553 unique syllables
90
+ - 48 unique phonemes
91
+
92
+ For more detailed analysis see the python notebook provided [here](https://gitlab.com/smc/msc-reviewed-speech/-/blob/master/analysis/EDA.ipynb)
93
 
94
+ ### Supported Tasks and Leaderboards
95
+
96
+ Automatic Speech Recognition system development, gender and age identification of speakers
97
+
98
  ### Languages
99
 
100
+ Malayalam
101
 
102
  ## Dataset Structure
103
 
104
+ - file_name
105
+ - speechid
106
+ - speaker_id
107
+ - review_score
108
+ - transcript
109
+ - category (optional speech category)
110
+ - speaker_gender (optionally self declared)
111
+ - speaker_age (optionally self declared)
112
+
113
  ### Data Instances
114
 
 
115
 
116
  ### Data Fields
117
 
 
118
 
119
  ### Data Splits
120
 
121
+ So specific Splits
122
 
123
  ## Dataset Creation
124
 
125
+ The speech data is collected from volunteer users who read and record their speech through a [web application](https://msc.smc.org.in) using their personal devices. The recorded speech is reviewed (upvote and downvote gives a score of +1 and -1 respectively) by other users. The review score is also published.
 
 
126
 
127
+ ### Curation Rationale
 
 
 
 
128
 
129
+ The recorded speech is reviewed (upvote and downvote gives a score of +1 and -1 respectively) by other users. The review score is also published.
130
 
131
+ ### Curation Rationale
132
 
133
+ Those speech samples with at least three positive reviews are included in this dataset.
134
 
135
+ ### Source Data
136
 
137
+ #### Initial Data Collection and Normalization
138
 
139
+ The speech data is collected from volunteer contributors who read and record their speech through a [web application](https://msc.smc.org.in). The users optionally provide name, age and gender. There is no further verification. Sentences to read out are curated by MSC Admin. The speech samples are reviewed by other users.
140
 
 
141
 
142
  ### Personal and Sensitive Information
143
 
144
+ Every speaker is identified by a unique alphanumeric id and age and gender are published if the speaker has voluntarily published them.
145
 
146
  ## Considerations for Using the Data
147
 
148
  ### Social Impact of Dataset
149
 
150
+ Read speech corpus, recorded in natural environments by the users.
 
 
 
 
 
 
 
 
151
 
 
152
 
153
  ### Dataset Curators
154
 
155
+ Kavya Manohar
156
 
157
  ### Licensing Information
158
 
159
+ CC-BY-SA 4.0
160
 
161
  ### Citation Information
162
 
 
163
 
164
  ### Contributions
165
 
166
+ http://msc.smc.org.in/