css10¶
Created by Kyubyong Park and Thomas Mulc
version |
1.0.0 |
license |
|
usage |
unrestricted |
languages |
zho, nld, fin, fra, deu, ell, hun, jpn, rus, spa |
format |
wav |
channel |
1 |
sampling rate |
22050 |
bit depth |
32 |
duration |
5 days 19:41:56.206122452 |
files |
63857, duration distribution: 1.0 s |
repository |
audb-public |
Description¶
CSS10 is a collection of single speaker speech data for 10 languages. Each of them consists of audio files recorded by a single volunteer and their aligned text sourced from LibriVox. The dataset contains: 6h for Chinese, 13h for Dutch, 10h for Finnish, 19h for French, 16h for German, 4h for Greek, 10h for Hungarian, 15h for Japanese, 21h for Russian, 24h for Spanish.
Tables¶
Click on a row to toggle a preview.
ID |
Type |
Columns |
---|---|---|
files |
filewise |
transcription, normalized-transcription, language, speaker |
Schemes¶
ID |
Dtype |
Labels |
---|---|---|
language |
str |
chinese, dutch, finnish, french, german, greek, hungarian, japanese, russian, spanish |
normalized-transcription |
str |
|
speaker |
str |
Bart de Leeuw, Diana Majlinger, Gilles G. Le Blanc, Harri Tapani Ylilammi, Hokuspokus, Jing Li, Mark Chulsky, Rapunzelina, Tux, ekzemplaro |
transcription |
str |