EfficientNet-Absolute Zero for Continuous Speech Keyword Spotting

31 Dec 2020  ·  Amir Mohammad Rostami, Ali Karimi, Mohammad Ali Akhaee ·

Keyword spotting is a process of finding some specific words or phrases in recorded speeches by computers. Deep neural network algorithms, as a powerful engine, can handle this problem if they are trained over an appropriate dataset. To this end, 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. The continuous speech synthesis method proposed to made FKD usable in the practical application which works with continuous speeches. Besides, we proposed a lightweight architecture called EfficientNet-A0 (absolute zero) by applying the compound scaling method on EfficientNet-B0 for keyword spotting task. Finally, the proposed architecture is evaluated with various models. It is realized that EfficientNet-A0 and Resnet models outperform other models on this dataset.

PDF Abstract

Datasets


Introduced in the Paper:

FKD

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Keyword Spotting CSS FKD EfficientNet-A0 + SA + TL Accuracy 76.84 # 1
Keyword Spotting FKD EfficientNet-A0 + SA + TL Accuracy 95.83 # 2
Keyword Spotting FKD Res26 Accuracy 95.88 # 1

Methods