Datasets:

YuDZEN
/

korebaju_corpus

speaker stringclasses 21 values	duration float64 0.28 2.4	gender stringclasses 2 values	text stringclasses 758 values
COE_13	0.416	M	o*a
COE_13	0.356	M	oʔa
COE_13	0.337	M	oʔa
COE_13	0.924	M	pinimɨ̃
COE_13	0.675	M	pinimõ
COE_13	0.94	M	pinimɨ̃
COE_13	0.646	M	kʰo*ɨme
COE_13	1.15	M	kʰoɨme
COE_13	0.835	M	kʰoɨme
COE_13	0.542	M	wese
COE_13	0.573	M	wese
COE_13	0.432	M	ɲjaˀseˀ
COE_13	0.547	M	ɲjaˀse
COE_13	0.562	M	ɲjaˀseˀ
COE_13	0.736	M	cimo
COE_13	1.066	M	cimo
COE_13	1.348	M	cimo
COE_13	0.643	M	makaɾɨ
COE_13	0.696	M	makaɾɨ
COE_13	0.626	M	makaɾɨ
COE_13	0.816	M	siuɲjɨ
COE_13	0.772	M	siuɲjɨ
COE_13	0.821	M	siuɲjɨ
COE_13	0.865	M	coˀcomɨ̃
COE_13	0.663	M	coˀcomẽ
COE_13	0.778	M	coˀcomɨ̃
COE_13	0.652	M	ɨ*ɨmɨ̃
COE_13	0.607	M	ɨ*ɨmɨ̃
COE_13	0.547	M	ɨ*ɨmɨ̃
COE_13	0.75	M	iɦace
COE_13	0.663	M	iɦace
COE_13	0.622	M	iɦace
COE_13	0.903	M	ca*o
COE_13	1.154	M	ca*o
COE_13	0.652	M	ca*o
COE_13	1.029	M	kaimɨ̃
COE_13	0.895	M	kaimɨ̃
COE_13	0.854	M	kaimɨ̃
COE_13	0.645	M	maɲa
COE_13	0.587	M	maɲa
COE_13	0.625	M	maɲa
COE_13	0.975	M	casamɨ̃
COE_13	1.124	M	cɨcɨ
COE_13	0.685	M	cɨˀcɨ
COE_13	0.715	M	cɨˀcɨ
COE_13	0.938	M	ka
COE_13	0.413	M	ka
COE_13	0.649	M	ka
COE_13	0.884	M	tʰeˀteɨ
COE_13	1.12	M	teteɨ
COE_13	0.703	M	tʰeteɾɨ
COE_13	0.578	M	βito
COE_13	0.529	M	βito
COE_13	0.512	M	βito
COE_13	0.533	M	cio
COE_13	0.623	M	cio
COE_13	0.616	M	cio
COE_13	0.724	M	koɨo
COE_13	0.59	M	koɨo
COE_13	0.534	M	koɨo
COE_13	0.684	M	ca*i
COE_13	0.896	M	ca*i
COE_13	0.615	M	sɨ̃sɨ
COE_13	0.577	M	sɨ̃sɨ
COE_13	0.512	M	sɨ̃sɨ
COE_13	0.639	M	ca*i
COE_13	1.549	M	poɦace
COE_13	0.886	M	poɦace
COE_13	1.234	M	poɦace
COE_13	0.968	M	casamɨ̃
COE_13	0.877	M	peumɨ̃
COE_13	0.629	M	peˀumɨ̃
COE_13	0.597	M	peumɨ̃
COE_13	0.815	M	casamɨ̃
COE_13	0.561	M	βeβɨ
COE_13	0.636	M	βeβɨ
COE_13	0.551	M	βeβɨ
COE_13	0.727	M	siˀswemɨ̃
COE_13	0.674	M	siˀswemɨ̃
COE_13	0.681	M	siˀswemɨ̃
COE_13	0.554	M	hena
COE_13	0.566	M	hena
COE_13	0.581	M	hena
COE_13	0.884	M	ukupiamɨ̃
COE_13	0.979	M	ceˀcome
COE_13	1.131	M	ceˀcome
COE_13	0.735	M	caˀcomẽ
COE_13	0.89	M	kʰweˀkomɨ̃
COE_13	0.679	M	kweˀkomɨ̃
COE_13	0.84	M	kweˀkomɨ̃
COE_13	0.828	M	ukupia*mɨ̃
COE_13	0.537	M	ha*a
COE_13	0.55	M	h*a
COE_13	0.55	M	ha*a
COE_13	0.687	M	βeβeɲɨ̃
COE_13	0.705	M	βeβeɲɨ̃
COE_13	0.664	M	βeβeɲɨ̃
COE_13	0.821	M	ukupiamɨ̃
COE_13	0.727	M	onona
COE_13	0.624	M	onona

End of preview. Expand in Dataset Viewer.

This is a Korebaju language dataset under construction.

check https://github.com/YuDZEN/korebaju-ASR for the Kaldi and MFA project for the same project

Audio Data

Audio data is in the audio folder. The audio data is in the wav format.

Transcription Data

Transcription data is test.jsonl and train.jsonl in the racine folder. The transcription data is in the jsonl format.

Data Split

The data is split into train and test sets. The train set contains 90% of the data and the test set contains 10% of the data.

data augmentation

We used data augmentation in order to increase the diversity of our training data without actually collecting new data (we have a very small dataset). By applying noise and volume, we can generate some training samples, which are named by "_NOISE" or "_VOLUME" after speakers' names.

Downloads last month: 181

Size of downloaded dataset files:

283 MB

Size of the auto-converted Parquet files:

271 MB

Number of rows:

7,977