JIBO Kids Corpus Transcripts are word and phonetic transcripts of speech in audio recordings of research interactions between children and a social robot (The JIBO Kids Corpus). Annotations are of structured and semi-structured speech from children from pre-kindergarten through grade 1.

The JIBO Kids Corpus Transcripts data shared here includes only transcripts. The corresponding audio can be downloaded by following the instructions in the Jibo_Kids github repository.

These transcripts were produced as part of a larger effort towards improved child speech recognition. Check out the Project Summary to learn more about the On Top of Pasketti Challenge, new state-of-the-art child ASR models released on Hugging Face, and all 520 hours of transcribed child speech collected as part of the project.

Permissions and licensing

License

Creative Commons Attribution 4.0

Key restrictions

Usage of this dataset should propertly cite both the DrivenData-generated transcripts and the original JIBO Kids Corpus.

JIBO Kids Corpus citations:

balaji1312. (2026). balaji1312/Jibo_Kids: JIBO Kids Corpus v1.0.0 (Version v1.0.0) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.19604408

Shankar, N. B., Afshan, A., Johnson, A., Mahapatra, A., Martin, A., Ni, H., Park, H. W., Perez, M. Q., Yeung, G., Bailey, A., Breazeal, C., & Alwan, A. (2024). The JIBO Kids Corpus: A speech dataset of child-robot interactions in a classroom environment. JASA Express Letters, 4(11), Article 115201. https://doi.org/10.1121/10.0034195

DrivenData-generated transcripts citation:

DrivenData. (2026). JIBO Kids Corpus Transcripts [Data set]. K-12 AI Infrastructure Program. https://platform.k12-ai-infrastructure.org/datasets/jibo

How to access the data

This dataset is available to anyone. To access the data, head to the Data download tab and follow the instructions provided.