FKD

Football Keywords Dataset

Introduced in EfficientNet-Absolute Zero for Continuous Speech Keyword Spotting2020

About this Dataset

The football keyword dataset (FKD), as a new keyword spotting dataset in Persian, is collected with crowdsourcing. This dataset contains nearly 31000 samples in 18 classes.

Dataset Variants

FKD

Papers1

EfficientNet-Absolute Zero for Continuous Speech Keyword Spotting

The continuous speech synthesis method is proposed to made FKD usable in the practical application which works with continuous speeches, and a lightweight architecture called EfficientNet-A0 (absolute zero) is proposed by applying the compound scaling method on E efficientNet-B0 for keyword spotting task.

Tasks

EDIT

Keyword Spotting Speaker Recognition Keyword Spotting CSS

Similar Datasets

MNIST

CelebA

GLUE

Statistics

Papers

1

Tasks

96

Introduced

2020

Languages

Iranian Persian