RawTFNet: A Lightweight CNN Architecture for Speech Anti-spoofing

TL;DR

RawTFNet is a lightweight CNN model for speech anti-spoofing that captures detailed features efficiently, achieving performance comparable to state-of-the-art models with less computational cost.

Contribution

Introduces RawTFNet, a novel lightweight CNN architecture that processes audio features along time and frequency for effective anti-spoofing.

Findings

Achieves comparable performance to state-of-the-art models on ASVspoof datasets.

Uses fewer computational resources than transformer-based models.

Demonstrates effective feature extraction for synthetic speech detection.

Abstract

Automatic speaker verification (ASV) systems are often affected by spoofing attacks. Recent transformer-based models have improved anti-spoofing performance by learning strong feature representations. However, these models usually need high computing power. To address this, we introduce RawTFNet, a lightweight CNN model designed for audio signals. The RawTFNet separates feature processing along time and frequency dimensions, which helps to capture the fine-grained details of synthetic speech. We tested RawTFNet on the ASVspoof 2021 LA and DF evaluation datasets. The results show that RawTFNet reaches comparable performance to that of the state-of-the-art models, while also using fewer computing resources. The code and models will be made publicly available.