Distinguishing AI-Generated Music from Edited Audio as a Hard-Negative Robustness Task

• URL: http://arxiv.org/abs/2608.14916v1
• Date: Fri, 14 Aug 2026 22:03:50 GMT
• Title: Distinguishing AI-Generated Music from Edited Audio as a Hard-Negative Robustness Task
• Abstract summary: Real-world uploads are often remixed, re-encoded, pitch-shifted, or otherwise edited.<n> edited versions form a difficult negative class: they are not generated by AI, yet they may introduce spectral artifacts that resemble synthetic audio fingerprints.<n>We study this problem as a hard-negative setting for AI-generated music detection, focusing on AI-generated and edited variants derived from the same anchor songs.
• Score: 39.146761527401424
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: AI-generated music detectors are commonly evaluated against original songs, but real-world uploads are often remixed, re-encoded, pitch-shifted, or otherwise edited. These edited versions form a difficult negative class: they are not generated by AI, yet they may introduce spectral artifacts that resemble synthetic audio fingerprints. We study this problem as a hard-negative robustness setting for AI-generated music detection, focusing on AI-generated and edited variants derived from the same anchor songs. We compile a YouTube-based dataset of AI, edited, and original variants, using the original tracks only as references, and train a binary AI versus edited detector. Audio is processed as 10-second clips and passed as raw waveforms to a pretrained PaSST spectrogram transformer. To reduce leakage, all splits are performed by anchor song. On the held-out test set, the final video-level system achieves 0.811 balanced accuracy. At clip level, AI-generated clips reach an F1-score of 0.836, while edited clips reach a lower F1-score of 0.720. The results suggest that AI-generated music retains detectable fingerprint-like spectral cues beyond ordinary editing, but the lower edited-class F1-score shows that these cues can still overlap with artifacts from edited audio. Grad-CAM visualizations are used to inspect whether high-confidence predictions rely on localized time-frequency regions.

Related papers

• How Much AI Is in This Track? Quantifying the Proportion of AI-Generated Stems in Hybrid Music Mixtures [39.04964205178628]
  Current AI music detection systems are binary, treating tracks as either fully AI or fully human.<n>We reformulate AI music detection as a regression problem on a continuous AI energy ratio, alpha in [0, 1].<n>We train a similar CNN-based model for regression of alpha, achieving MAE = 0.076 and R2 = 0.85 on heldout mixtures from the same pipeline.
  arXiv  Detail & Related papers  (2026-08-07T14:46:06Z)
• FIGMA: Towards FIne-Grained Music retrievAl [65.98380295254817]
  Retrieving music using natural language descriptions has improved with contrastive audio-text models such as CLAP, but current systems remain limited to coarse semantic queries.<n>We show that this limitation stems from the contrastive learning objective itself: despite being trained on long captions, CLAP-based models effectively utilize only the first few tokens.<n>We propose FIGMA, a multi-view contrastive architecture that addresses this limitation by jointly optimizing global audio-text alignment and frame-level, token-wise alignment.
  arXiv  Detail & Related papers  (2026-06-04T18:05:39Z)
• Operation-Guided Progressive Human-to-AI Text Transformation Benchmark for Multi-Granularity AI-Text Detection [49.41415417307568]
  OpAI-Bench is an operation-guided benchmark for studying progressive human-to-AI text transformation.<n>We show that AI-text detectability is governed not only by the proportion of AI-edited content, but also by edit operation, domain, and cumulative revision history.
  arXiv  Detail & Related papers  (2026-06-04T17:58:05Z)
• AI-Generated Music Detection in Broadcast Monitoring [1.1937220268355655]
  We introduce AI-OpenBMAT, the first dataset tailored to broadcast-style AI-music detection.<n>It contains 3,294 one-minute audio excerpts that follow the duration patterns and loudness relations of real television audio.<n>We benchmark a CNN baseline and state-of-the-art SpectTTTra models to assess SNR and duration robustness, and evaluate on a full broadcast scenario.
  arXiv  Detail & Related papers  (2026-02-06T16:08:01Z)
• Skyra: AI-Generated Video Detection via Grounded Artifact Reasoning [66.51617619673587]
  We present Skyra, a specialized large language model (MLLM) that identifies human-perceivable visual artifacts in AI-generated videos.<n>To support this objective, we construct ViF-CoT-4K for Supervised Fine-Tuning (SFT), which represents the first large-scale AI-generated video dataset with fine-grained human annotations.<n>We then develop a two-stage training strategy that systematically enhances our model's artifact's-temporal perception, explanation capability, and detection accuracy.
  arXiv  Detail & Related papers  (2025-12-17T18:48:26Z)
• Segment Transformer: AI-Generated Music Detection via Music Structural Analysis [1.7034813545878587]
  We aim to improve the accuracy of AIGM detection by analyzing the structural patterns of music segments.<n>Specifically, to extract musical features from short audio clips, we integrated various pre-trained models.<n>For long audio, we developed a segment transformer that divides music into segments and learns inter-segment relationships.
  arXiv  Detail & Related papers  (2025-09-10T04:56:40Z)
• DAVID-XR1: Detecting AI-Generated Videos with Explainable Reasoning [58.70446237944036]
  DAVID-X is the first dataset to pair AI-generated videos with detailed defect-level, temporal-spatial annotations and written rationales.<n>We present DAVID-XR1, a video-language model designed to deliver an interpretable chain of visual reasoning.<n>Our results highlight the promise of explainable detection methods for trustworthy identification of AI-generated video content.
  arXiv  Detail & Related papers  (2025-06-13T13:39:53Z)
• Detecting Music Performance Errors with Transformers [3.6837762419929168]
  Existing tools for music error detection rely on automatic alignment.<n>There is a lack of sufficient data to train music error detection models.<n>We present a novel data generation technique capable of creating large-scale synthetic music error datasets.
  arXiv  Detail & Related papers  (2025-01-03T07:04:20Z)
• SONICS: Synthetic Or Not -- Identifying Counterfeit Songs [0.16777183511743465]
  We introduce SONICS, a novel dataset for end-to-end Synthetic Song Detection (SSD)<n>We highlight the importance of modeling long-range temporal dependencies in songs for effective authenticity detection.<n>For long songs, our top-performing variant outperforms ViT by 8% in F1 score, is 38% faster, and uses 26% less memory.
  arXiv  Detail & Related papers  (2024-08-26T08:02:57Z)
• Fully Automated End-to-End Fake Audio Detection [57.78459588263812]
  This paper proposes a fully automated end-toend fake audio detection method. We first use wav2vec pre-trained model to obtain a high-level representation of the speech. For the network structure, we use a modified version of the differentiable architecture search (DARTS) named light-DARTS.
  arXiv  Detail & Related papers  (2022-08-20T06:46:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.