Seoul Corpus
The Seoul Corpus pairs spontaneous Korean speech with phonetic annotation.
Visit official source (opens in a new tab)Research use
Study connected speech and pronunciation in spontaneous Korean, or build a Korean adaptation evaluation with labeled segments.
Limitations
The label archive uses Praat TextGrid. Experiments should document the conversion from phonetic or word annotations to ASR scoring targets.
License and access
- Dataset license
- CC BY-NC 2.0 (opens in a new tab)
- Access status
- Direct access
Download the audio, label and documentation archives separately from OpenSLR.