The dataset is currently empty. Upload or create new data files. Then, you will be able to explore them in the Dataset Viewer.

Parakeet Dictation Cleanup Eval

Status (2026-10-16): placeholder. No clips recorded yet.

Evaluation only. Never train on this set.

Planned contents: 40 spontaneous dictations by one speaker, recorded from fictional scenario cards (recording/scenarios.json in https://github.com/danielrosehill/ASR-Post-Processor-Model), transcribed by stock nvidia/parakeet-tdt-0.6b-v2, and hand-cleaned at both control settings ([Layers: 1] and [Layers: 1,2]). The cast and details on the cards are fictional, so the set carries no real personal data.

Downloads last month
36