Korean Speech Dataset

When this dataset fits

Korean Speech Dataset is a Korean Speech Dataset from AI Hub. It is closer to speech recognition than to read-speech TTS. License label AI Hub 이용안내. Size text is “4TB”, copied from the host.

Task / category

Speech · Speech · sign-language, video

Language

Korean

Format · Size

Audio · 4TB (12K)

License

AI Hub terms(Catalog verified) · License document

Source & original link

AI Hub · URL not verified

Verification status

Source: Unknown · License: Verified · Last checked: 2026-09-03

Korean Speech Dataset: Korean Speech Dataset from AI Hub. License license on original source. Open the original link to download.

Korean Speech Dataset (2021). AI Hub. license label: AI Hub terms catalog commercial mark: conditional Original URL is not stored on this catalog row. GearDel catalog: https://geardel.com/en/datasets/aihub-sign-language. Last checked: 2026-09-03. Author names are not stored in the GearDel catalog and are not invented here.

Differences below use catalog fields only: license, commercial mark, provider, size text, and year.

• Korean Speech DatasetKorean Speech Dataset · AI HubViewSize text differs (4TB vs 800hours). Those strings are copied from the hosts.
• Korean Speech DatasetKorean Speech Dataset · AI HubViewSize text differs (4TB vs 120GB). Those strings are copied from the hosts.
• Korean Speech Synthesis DatasetKorean Speech Synthesis Dataset · AI HubViewSize text differs (4TB vs 100hours). Those strings are copied from the hosts. Catalog years differ (2021 vs 2020).
• TTSKorean Speech Synthesis Dataset · AI HubViewSize text differs (4TB vs 300hours). Those strings are copied from the hosts.