Datasetpublic

iclr2027anon/animalspeak-pseudovox

AnimalSpeak Pseudovox This dataset contains 346,907 short, silence-trimmed animal vocalization clips from the train-unseen split. WAV audio is included in data-20k/train-*.tar. Quick start from datasets import load_dataset metadata = load_dataset("iclr2027anon/animalspeak-pseudovox", split="train") Download and extract the TAR shards from data-20k/. Load a row's WAV from audio/<audio_name>. Metadata Column Meaning id Clip ID: pseudovox_… See the full description on the dataset page: https://huggingface.co/datasets/iclr2027anon/animalspeak-pseudovox.

sourceHugging Facecc-by-nc-4.0updated 11h agoView on Hugging Face

0likes54downloads

Dataset Card

AnimalSpeak Pseudovox

This dataset contains 346,907 short, silence-trimmed animal vocalization clips from the train-unseen split. WAV audio is included in data-20k/train-*.tar.

Quick start

python

from datasets import load_dataset

metadata = load_dataset("iclr2027anon/animalspeak-pseudovox", split="train")

Download and extract the TAR shards from data-20k/. Load a row's WAV from audio/<audio_name>.

Metadata

Column	Meaning
`id`	Clip ID: `pseudovox_` followed by 32 hexadecimal characters.
`audio_name`	WAV filename inside a TAR shard's `audio/` directory.
`duration_s`	Clip duration in seconds.
`species_scientific_gbif`	Source species label.
`custom_classifier_species`, `custom_classifier_max_p`	Predicted species and classifier confidence.
`birdnet_confidence`	BirdNET confidence.
`source_url`	Original recording page, where available.

IDs identify rows; filenames locate audio. ROOTS references these exact IDs in source_audio_ids.

License

Source licenses and attribution requirements apply to the audio. See the repository's license metadata and original recording sources.