# Half-Truth Audio Detection and Localisation: A Lightweight Cross-Attentive Architecture and a Cross-Corpus Diagnostic Study

arXiv: https://arxiv.org/abs/2605.29531 · HTML: https://mlsanity.com/paper/2605.29531/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 27 references that are in the map.
- First posted on arXiv 28 May 2026.

## Builds on (most cited first, 17 of 17)

- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [AST: Audio Spectrogram Transformer](https://mlsanity.com/paper/2104.01778/) (2021)
- [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://mlsanity.com/paper/2111.09296/) (2021)
- [BigVGAN: A Universal Neural Vocoder with Large-Scale Training](https://mlsanity.com/paper/2206.04658/) (2022)
- [ASVspoof 2021: accelerating progress in spoofed and deepfake speech detection](https://mlsanity.com/paper/2109.00537/) (2021)
- [Does Audio Deepfake Detection Generalize?](https://mlsanity.com/paper/2203.16263/) (2022)
- [The PartialSpoof Database and Countermeasures for the Detection of Short Fake Speech Segments Embedded in an Utterance](https://mlsanity.com/paper/2204.05177/) (2022)
- [Half-Truth: A Partially Fake Audio Detection Dataset](https://mlsanity.com/paper/2104.03617/) (2021)
- [Speech is Silver, Silence is Golden: What do ASVspoof-trained Models Really Learn?](https://mlsanity.com/paper/2106.12914/) (2021)
- [An Initial Investigation for Detecting Partially Spoofed Audio](https://mlsanity.com/paper/2104.02518/) (2021)
- [SpoofCeleb: Speech Deepfake Detection and SASV In The Wild](https://mlsanity.com/paper/2409.17285/) (2024)
- [Warning: Humans Cannot Reliably Detect Speech Deepfakes](https://mlsanity.com/paper/2301.07829/) (2023)
- [The Impact of Silence on Speech Anti-Spoofing](https://mlsanity.com/paper/2309.11827/) (2023)
- [Comprehensive Layer-wise Analysis of SSL Models for Audio Deepfake Detection](https://mlsanity.com/paper/2502.03559/) (2025)
- [Defense Against Adversarial Attacks on Audio DeepFake Detection](https://mlsanity.com/paper/2212.14597/) (2022)
- [MFAAN: Unveiling Audio Deepfakes with a Multi-Feature Authenticity Network](https://mlsanity.com/paper/2311.03509/) (2023)

## Abstract (arXiv)

Partially manipulated (half-truth) speech, where a short synthesised segment is spliced into an otherwise genuine utterance, is a harder and more realistic forensic threat than the fully synthesised deepfakes that dominate the literature. We present CAFNet, a lightweight (576K-parameter, 2.24 MB) cross-attentive architecture that fuses MFCC, LFCC, and Chroma-STFT features to jointly classify audio as real, fully fake, or half-truth, and regress the temporal boundaries of the synthesised region, at approximately 14 ms CPU latency. A component ablation shows cross-attention fusion is CAFNet's most load-bearing component; a deeply supervised auxiliary classification head from earlier iterations is not, and removing it improves every in-domain metric under 3-seed replication with substantially lower variance. On MLADDC T2+T3 the model reaches 97.55%±0.69% ternary accuracy and 0.037 s boundary mean absolute error (MAE), to our knowledge, the first reported continuous splice- boundary localisation result on this benchmark. Zero-shot evaluation on two independent benchmarks shows transfer is capability- and corpus-dependent rather than uniform: on Half-Truth Audio Detection dataset (HAD), detection recall reaches 84.9% and ternary classification resolves half-truth correctly on half of true half-truth clips (50.4%), while on PartialSpoof, binary detection stays near chance (AUC 0.5544). We treat this asymmetry, not a single generalization verdict, as the finding. HAD localisation improves in absolute terms but degrades in relative terms, since in-domain localisation improved faster. An architectural change validated purely in-domain thus shifted the cross-corpus transfer profile, evidence that cross-corpus evaluation should accompany, not follow, in-domain architecture decisions.