Datasets:

luvox-ai
/

smart-vad-done-preview

Tasks:

Audio Classification

Modalities:

Formats:

Size:

Tags:

voice-activity-detection

Libraries:

License:

Dataset card Data Studio Files Files and versions

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

Smart-VAD DONE Samples Preview

Private preview export for Smart-VAD annotation samples.

Repository target: luvox-ai/smart-vad-done-preview
Rows: 48
Includes only annotations with annotation_status == "done"
draft and pending audio are intentionally excluded

The dataset uses an AudioFolder-compatible layout:

data/train-00000-of-00001.parquet
metadata.csv
audio/<audio_id>.<ext>
annotations/annotation_done.jsonl
speaker_registry.json
speaker_registry.jsonl
reports/validation_report.json

Important publish rule: future Parquet/Hugging Face exports must only use records where annotation_status == "done".

Downloads last month: 12

Total file size: