A Detection Algorithm for Audio Adversarial Examples in EI-Enhanced Automatic Speech Recognition

Table 3

Software environment.

(a)

Component	Version
Ubuntu	16.04
Python	3.5.2
Deep Speech code	0.4.1
Deep Speech model	0.4.1
CUDA	9.0
cuDNN	7.0

(b)

Python packages	Version
pandas	0.24.0
numpy	1.16.4
Keras	2.2.4
ds-ctcdecoder	0.4.1
tensoflow-gpu	1.12.0
scipy	1.4.1