Yixuan/cl-fad-ver0

This repo is used to host the configs and checkpoints used in our ICASSP2024 submission projects.

How to run

The codebase is work-in-progress.

Exp Description

For each model, there are four exp setups, as shown in the following table.

For example, exp1.1 refers to an experiment where:

• The genuine data comes from Librispeech.
• The training order is organized by attacker's training dataset's name.
• The training strategy is finetuning.

Genuine Data	Librispeech		MLAAD
Task Order	By Dataset	Random	Random
Finetune	exp1.1	exp1.2	exp1.4
One-Class Softmax	exp2.1	exp2.2	exp2.4
Replay	exp3.1	exp3.2	exp3.4
LwF	exp4.1	exp4.2	exp4.4