DBD-research-group/beans_sc

---
dataset_info:
features:
- name: 'Unnamed: 0'
dtype: int64
- name: path
dtype: audio
- name: label
dtype: string
splits:
- name: train
num_bytes: 2681260515. 414
num_examples: 84843
- name: valid
num_bytes: 312665460.466
num_examples: 9981
- name: test
num_bytes: 344947874.825
num_examples: 11005
- name: train_low
num_bytes: 26658119.0
num_examples: 849
download_size: 3024419370
dataset_size: 3365531969.705
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*
- split: train_low
path: data/train_low-*
---
# Dataset Card for "beans_sc"

## Dataset Description

**Paper:** https://doi.org/10.48550/arXiv.1804.03209

## Dataset Summary

This dataset contains single-word utterances covering 35 English words including digits, commands, and others, spoken by multiple speakers with splits and preprocessing like described in BEANS. It is used for **classification** tasks and is an **auxiliary** dataset in BEANS.

## Data Splits

| train | train_low | valid | test |
|------|-----| ------ | ------ |
| 84843 | 849 | 9981 | 11005 |