The dataset viewer is not available for this split.
Parquet error: Scan size limit exceeded: attempted to read 592700208 bytes, limit is 300000000 bytes
Make sure that
1. the Parquet files contain a page index to enable random access without loading entire row groups2. otherwise use smaller row-group sizes when serializing the Parquet files
Error code: TooBigContentError
Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
TurnBench baseline probs
Per-frame end-of-turn and interruption scores for the threshold-based
TurnBench baselines (dev and test splits), one directory per baseline,
mirroring baselines/<name>/ in
cmu-sesame/turnbench.
These files are the input to the paper's operating-point protocol: a common
commit rule converts scores to commit times, and each baseline's operating
point is the highest-recall threshold whose dev false-positive rate stays
within the 0.1 budget. The benchmark code fetches them at a pinned revision
(eval/probs.py). Contains no audio and no labels.
- Downloads last month
- 135