ProgramComputer/avspeech-visual-audio
收藏Hugging Face2026-02-20 更新2026-03-29 收录
下载链接：
https://hf-mirror.com/datasets/ProgramComputer/avspeech-visual-audio
下载链接
链接失效反馈官方服务：
资源简介：
---
configs:
- config_name: default
data_files:
- split: train
path: train/*.parquet
- split: test
path: test/*.parquet
---
# AVSpeech Video + Audio
A restructured subset of the [AVSpeech](https://looking-to-listen.github.io/avspeech/) dataset with separated media streams and derived identifiers.
- `clip_id`: unique identifier per clip, derived as `{youtube_id}_{start:.3f}_{end:.3f}` from the original AVSpeech CSV columns.
- `avspeech_metadata`: JSON string with the original AVSpeech row fields (`youtube_id`, `start_sec`, `end_sec`, `x_center`, `y_center`).
- `video`: video-only stream (no audio), stream-copied without re-encoding.
- `audio`: audio-only stream, stream-copied without re-encoding.
## Citation
If you use this dataset, please cite the original AVSpeech paper:
```bibtex
@inproceedings{ephrat2018looking,
title={Looking to listen at the cocktail party: A speaker-independent audio-visual model for speech separation},
author={Ephrat, Ariel and Mosseri, Inbar and Lang, Oran and Dekel, Tali and Wilson, Kevin and Hassidim, Avinatan and Freeman, William T and Rubinstein, Michael},
booktitle={ACM SIGGRAPH 2018},
year={2018}
}
```
提供机构：
ProgramComputer
