HQ-MPSD: A Multilingual Artifact-Controlled Benchmark for Partial Deepfake Speech Detection
By: Menglu Li , Majd Alber , Ramtin Asgarianamiri and more
Detecting partial deepfake speech is challenging because manipulations occur only in short regions while the surrounding audio remains authentic. However, existing detection methods are fundamentally limited by the quality of available datasets, many of which rely on outdated synthesis systems and generation procedures that introduce dataset-specific artifacts rather than realistic manipulation cues. To address this gap, we introduce HQ-MPSD, a high-quality multilingual partial deepfake speech dataset. HQ-MPSD is constructed using linguistically coherent splice points derived from fine-grained forced alignment, preserving prosodic and semantic continuity and minimizing audible and visual boundary artifacts. The dataset contains 350.8 hours of speech across eight languages and 550 speakers, with background effects added to better reflect real-world acoustic conditions. MOS evaluations and spectrogram analysis confirm the high perceptual naturalness of the samples. We benchmark state-of-the-art detection models through cross-language and cross-dataset evaluations, and all models experience performance drops exceeding 80% on HQ-MPSD. These results demonstrate that HQ-MPSD exposes significant generalization challenges once low-level artifacts are removed and multilingual and acoustic diversity are introduced, providing a more realistic and demanding benchmark for partial deepfake detection. The dataset can be found at: https://zenodo.org/records/17929533.
Similar Papers
Multilingual Source Tracing of Speech Deepfakes: A First Benchmark
Audio and Speech Processing
Finds who made fake voices, even in other languages.
Multilingual Dataset Integration Strategies for Robust Audio Deepfake Detection: A SAFE Challenge System
Audio and Speech Processing
Finds fake voices in recordings.
Multimodal Zero-Shot Framework for Deepfake Hate Speech Detection in Low-Resource Languages
Sound
Finds hate speech in fake voices, even new ones.