Datasets:

formospeech
/

hakkadict_moe_word

Modalities:

Formats:

Size:

Libraries:

Dataset card Data Studio Files Files and versions

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

TRAIN

Subset	lang_group	hours	n_utts	n_chars	secs/utt	chars/sec
Hakka_Dapu	客語_大埔	7.40	11,603	25,933	2.30	0.97
Hakka_Hailu	客語_海陸	4.98	11,415	25,531	1.57	1.42
Hakka_NanSixian	客語_南四縣	5.81	11,679	26,292	1.79	1.26
Hakka_Raoping	客語_饒平	3.81	11,390	25,498	1.20	1.86
Hakka_Sixian	客語_四縣	4.41	11,397	25,532	1.39	1.61
Hakka_Zhaoan	客語_詔安	5.19	9,946	21,271	1.88	1.14
Total	-	31.60	67,430	150,057	1.69	1.32

Downloads last month: 39

Collection including formospeech/hakkadict_moe_word

Taiwanese Hakka for ASR (Kaldi Init.)

Speech for ASR (Kaldi Init.) training • 14 items • Updated 10 days ago