Audio deepfake detection is well-studied as a binary problem, but partially manipulated speech, where a short synthesised segment is spliced into an otherwise genuine utterance, poses a harder and more realistic threat. Detecting such half-truth audio requires not only distinguishing it from real and fully fake speech, but also localising where the manipulation occurs. We present CAFNet, a 576k-parameter architecture that addresses both tasks jointly: it performs ternary classification (real, fully-fake, or half-truth) and regresses the temporal boundaries of the synthesised region in a single forward pass. CAFNet fuses Mel-Frequency Cepstral Coefficient (MFCC), Linear-Frequency Cepstral Coefficient (LFCC), and Chroma Short-Time Fourier Transform (Chroma-STFT) features through parallel depthwise-separable convolution branches with cross-attention, followed by a Bidirectional Long Short-Term Memory (BiLSTM) regression head for boundary prediction. On the combined Multi-Lingual Audio Deepfake Detection Corpus (MLADDC) T2+T3 test set, CAFNet achieves 92.71% accuracy and macro Area Under the Curve (AUC) of 0.9910, with boundary localisation Mean Absolute Error (MAE) of 0.075s and a median error of 0.052s. On binary detection, it achieves 96.76% accuracy and 3.20% Equal Error Rate (EER), outperforming fine-tuned XLS-R 300M (78.31%) and AST 87M (93.03%) at over 500 times fewer parameters. A cross-dataset study further shows that standard fine-tuning collapses cross-domain representations even under reduced backbone learning rates.

翻译：音频深度伪造检测作为二分类问题已被广泛研究，但部分篡改语音——即短合成片段嵌入真实话语的音频——构成更困难且更具现实威胁的挑战。检测此类半真音频不仅需要区分真实语音、完全伪造语音，还需定位篡改发生的时间位置。我们提出CAFNet（576k参数架构），通过单次前向传播联合执行两项任务：三分类（真实、完全伪造或半真）以及合成区域时间边界的回归。CAFNet通过并行深度可分离卷积分支与交叉注意力机制，融合梅尔频率倒谱系数（MFCC）、线性频率倒谱系数（LFCC）和色度短时傅里叶变换（Chroma-STFT）特征，随后接入双向长短期记忆（BiLSTM）回归头进行边界预测。在联合多语言音频深度伪造检测语料库（MLADDC）T2+T3测试集上，CAFNet达到92.71%的准确率和0.9910的宏曲线下面积（AUC），边界定位平均绝对误差（MAE）为0.075秒，中位误差为0.052秒。在二分类检测中，它以96.76%的准确率和3.20%的等错误率（EER）超越微调后的XLS-R 300M（78.31%）和AST 87M（93.03%），参数量减少超过500倍。跨数据集研究进一步表明，即使降低骨干网络学习率，标准微调仍会导致跨域表示崩溃。