面向合成语音检测的弹性正交权重修正连续学习算法.pdfVIP

计算机研究与发展

Journal

of

Computer

Research

and

Development62（2）：336−345，2025

面向合成语音检测的弹性正交权重修正连续学习算法

1,2131

张晓辉易江燕陶建华周俊佐

1

（中国科学院自动化研究所北京100190）

2

（北京交通大学计算机与信息技术学院北京100044）

3

（清华大学自动化系北京100084）

）

ElasticOrthogonalWeightModificationContinualLearningAlgorithminthe

ContextofSyntheticSpeechDetection

1,2

1

3

1

Zhang

Xiaohui,

Yi

Jiangyan,

Tao

Jianhua,

and

Zhou

Junzuo

1

（InstituteofAutomation,ChineseAcademyofSciences,Beijing100190）

2

（SchoolofComputerandInformationTechnology,BeijingJiaotongUniversity,Beijing100044）

3

（DepartmentofAutomation,TsinghuaUniversity,Beijing100084）

AbstractCurrently,

deep

learning

has

achieved

significant

success

in

the

field

of

synthetic

speech

detection.

However,

deep

models

commonly

attain

high

accuracy

on

test

sets

that

closely

match

their

training

distribution

but

exhibit

a

substantial

drop

in

accuracy

in

cross-dataset

scenarios.

To

enhance

the

generalization

capability

of

models

on

new

datasets,

they

are

often

fine-tuned

with

new

data,

but

this

leads

to

catastrophic

forgetting,

where

the

model’s

knowledge

learned

from

old

data

is

impaired,

resulting

in

deteriorated

performance

on

the

old

data.

Continuous

learning

is

a

prevalent

approach

to

mitigate

catastrophic

forgetting.

In

this

paper,

we

propose

a

continuous

learning

algorithm

called

elastic

orthogonal

weight

modification

(EOWM)

to

address

catastrophic

forgetting

for

synthetic

speech

detection.

EOWM

mitigates

knowledge

degradation

by

adjusting

the

direction

and

magnitude

of

parameter