datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
reachy-mini-emotions-library
Reachy Mini Emotions Library
Curated emotion recordings for the Reachy Mini robot, maintained by
Pollen Robotics. Each move is a JSON trajectory (head pose, antennas,
body yaw, sampled over time) paired with an Opus audio track.
Motion is sampled at 50 Hz; audio is mono Ogg/Opus (decoded natively by
the robot). Requires reachy_mini ≥ v1.8.4 (its move loader resolves
non-.wav audio sidecars).
File layout
Files live at the root of the dataset, named <emotion>.json +… See the full description on the dataset page: https://huggingface.co/datasets/pollen-robotics/reachy-mini-emotions-library.Reachy_RENS
Reachy RENS - Motor Noise Dataset
Crowdsourced recordings of Reachy Mini robot motor noise for training ego-noise suppression models.
Purpose
This dataset enables training of real-time motor noise cancellation models (DTLN) that can:
Remove motor whine from live audio streams
Improve audio analysis (beat detection, speech recognition)
Create cleaner livestream audio for robotics applications
Structure
data/
├── {user_hash_1}/
│ ├── background.wav # 15s… See the full description on the dataset page: https://huggingface.co/datasets/TwinPeaksTownie/Reachy_RENS.reachy-mini-emotions-library-50hz-opus
Reachy Mini Emotions Library — 50 Hz + Opus (compressed test build)
Compressed variant of the
Reachy Mini Emotions Library,
produced to evaluate a smaller on-disk / faster-to-upload move format. Same 81
moves, same choreography and audio content — only the encoding changed:
Motion re-sampled to 50 Hz (was 100 Hz) using the SDK's own
interpolation, with float precision rounded to 6 decimals
(1 micron / 1 microradian — physically lossless for this hardware).
Audio transcoded from… See the full description on the dataset page: https://huggingface.co/datasets/RemiFabre/reachy-mini-emotions-library-50hz-opus.reachy-mini-onboarding-movesReachy_RENS