Bob's Confetti: Phonetic Memorization Attacks in Music and Video Generation

• URL: http://arxiv.org/abs/2507.17937v3
• Date: Wed, 29 Oct 2025 17:29:43 GMT
• Title: Bob's Confetti: Phonetic Memorization Attacks in Music and Video Generation
• Authors: Jaechul Roh, Zachary Novack, Yuefeng Peng, Niloofar Mireshghallah, Taylor Berg-Kirkpatrick, Amir Houmansadr,
• Abstract summary: Generative AI systems for music and video commonly use text-based filters to prevent the regurgitation of copyrighted material.<n>We introduce Adversarial PhoneTic Prompting (APT), a novel attack that bypasses these safeguards by exploiting phonetic memorization.<n>We demonstrate that leading Lyrics-to-Song (L2S) models like SUNO and YuE regenerate songs with striking melodic and rhythmic similarity to their copyrighted originals when prompted with these altered lyrics.
• Score: 47.04195212078377
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Generative AI systems for music and video commonly use text-based filters to prevent the regurgitation of copyrighted material. We expose a fundamental flaw in this approach by introducing Adversarial PhoneTic Prompting (APT), a novel attack that bypasses these safeguards by exploiting phonetic memorization. The APT attack replaces iconic lyrics with homophonic but semantically unrelated alternatives (e.g., "mom's spaghetti" becomes "Bob's confetti"), preserving acoustic structure while altering meaning; we identify high-fidelity phonetic matches using CMU pronouncing dictionary. We demonstrate that leading Lyrics-to-Song (L2S) models like SUNO and YuE regenerate songs with striking melodic and rhythmic similarity to their copyrighted originals when prompted with these altered lyrics. More surprisingly, this vulnerability extends across modalities. When prompted with phonetically modified lyrics from a song, a Text-to-Video (T2V) model like Veo 3 reconstructs visual scenes from the original music video-including specific settings and character archetypes-despite the absence of any visual cues in the prompt. Our findings reveal that models memorize deep, structural patterns tied to acoustics, not just verbatim text. This phonetic-to-visual leakage represents a critical vulnerability in transcript-conditioned generative models, rendering simple copyright filters ineffective and raising urgent concerns about the secure deployment of multimodal AI systems. Demo examples are available at our project page (https://jrohsc.github.io/music_attack/).

Related papers

• VEIL: Jailbreaking Text-to-Video Models via Visual Exploitation from Implicit Language [25.38940067963429]
  Prior attacks on text-to-video (T2V) models typically add adversarial perturbations to obviously unsafe prompts.<n>We show that benign-looking prompts containing rich, implicit cues can induce T2V models to generate semantically unsafe videos.<n>We propose VEIL, a jailbreak framework that leverages T2V models' cross-modal associative patterns via a modular prompt design.
  arXiv  Detail & Related papers  (2025-11-17T08:31:43Z)
• Text2Lip: Progressive Lip-Synced Talking Face Generation from Text via Viseme-Guided Rendering [53.2204901422631]
  Text2Lip is a viseme-centric framework that constructs an interpretable phonetic-visual bridge.<n>We show that Text2Lip outperforms existing approaches in semantic fidelity, visual realism, and modality robustness.
  arXiv  Detail & Related papers  (2025-08-04T12:50:22Z)
• ThinkSound: Chain-of-Thought Reasoning in Multimodal Large Language Models for Audio Generation and Editing [52.33281620699459]
  ThinkSound is a novel framework that leverages Chain-of-Thought (CoT) reasoning to enable stepwise, interactive audio generation and editing for videos.<n>Our approach decomposes the process into three complementary stages: semantically coherent, interactive object-centric refinement through precise user interactions, and targeted editing guided by natural language instructions.<n> Experiments demonstrate that ThinkSound achieves state-of-the-art performance in video-to-audio generation across both audio metrics and CoT metrics.
  arXiv  Detail & Related papers  (2025-06-26T16:32:06Z)
• Double Entendre: Robust Audio-Based AI-Generated Lyrics Detection via Multi-View Fusion [11.060929679400667]
  We propose a multimodal, modular late-fusion pipeline that combines automatically transcribed lyrics and speech features capturing lyrics-related information within the audio.<n>Our method, DE-detect, outperforms existing lyrics-based detectors while also being more robust to audio perturbations.
  arXiv  Detail & Related papers  (2025-06-19T02:56:49Z)
• Shushing! Let's Imagine an Authentic Speech from the Silent Video [15.426152742881365]
  Vision-guided speech generation aims to produce authentic speech from facial appearance or lip motions without relying on auditory signals.<n>Despite recent progress, existing methods struggle to achieve unified cross-modal alignment across semantics, timbre, and emotional prosody from visual cues.<n>We introduce ImaginTalk, a novel cross-modal diffusion framework that generates faithful speech using only visual input.
  arXiv  Detail & Related papers  (2025-03-19T06:28:17Z)
• REFFLY: Melody-Constrained Lyrics Editing Model [50.03960548399128]
  This paper introduces REFFLY, the first revision framework for editing and generating melody-aligned lyrics.<n>We train the lyric revision module using our synthesized melody-aligned lyrics dataset.<n>To further enhance the revision ability, we propose training-frees aimed at preserving both semantic meaning and musical consistency.
  arXiv  Detail & Related papers  (2024-08-30T23:22:34Z)
• Synthetic Lyrics Detection Across Languages and Genres [4.987546582439803]
  Large language models (LLMs) to generate music content, particularly lyrics, has gained in popularity.<n>Previous research has explored content detection in various domains, but no work has focused on the text modality, lyrics, in music.<n>We curated a diverse dataset of real and synthetic lyrics from multiple languages, music genres, and artists.<n>We performed a thorough evaluation of existing synthetic text detection approaches on lyrics, a previously unexplored data type.<n>Following both music and industrial constraints, we examined how well these approaches generalize across languages, scale with data availability, handle multilingual language content, and perform on novel genres in few-shot settings
  arXiv  Detail & Related papers  (2024-06-21T15:19:21Z)
• Syllable-level lyrics generation from melody exploiting character-level language model [14.851295355381712]
  We propose to exploit fine-tuning character-level language models for syllable-level lyrics generation from symbolic melody. In particular, our method endeavors to incorporate linguistic knowledge of the language model into the beam search process of a syllable-level Transformer generator network.
  arXiv  Detail & Related papers  (2023-10-02T02:53:29Z)
• Unsupervised Melody-to-Lyric Generation [91.29447272400826]
  We propose a method for generating high-quality lyrics without training on any aligned melody-lyric data. We leverage the segmentation and rhythm alignment between melody and lyrics to compile the given melody into decoding constraints. Our model can generate high-quality lyrics that are more on-topic, singable, intelligible, and coherent than strong baselines.
  arXiv  Detail & Related papers  (2023-05-30T17:20:25Z)
• Unsupervised Melody-Guided Lyrics Generation [84.22469652275714]
  We propose to generate pleasantly listenable lyrics without training on melody-lyric aligned data. We leverage the crucial alignments between melody and lyrics and compile the given melody into constraints to guide the generation process.
  arXiv  Detail & Related papers  (2023-05-12T20:57:20Z)
• Noise2Music: Text-conditioned Music Generation with Diffusion Models [73.74580231353684]
  We introduce Noise2Music, where a series of diffusion models is trained to generate high-quality 30-second music clips from text prompts. We find that the generated audio is not only able to faithfully reflect key elements of the text prompt such as genre, tempo, instruments, mood, and era. Pretrained large language models play a key role in this story -- they are used to generate paired text for the audio of the training set and to extract embeddings of the text prompts ingested by the diffusion models.
  arXiv  Detail & Related papers  (2023-02-08T07:27:27Z)
• Re-creation of Creations: A New Paradigm for Lyric-to-Melody Generation [158.54649047794794]
  Re-creation of Creations (ROC) is a new paradigm for lyric-to-melody generation. ROC achieves good lyric-melody feature alignment in lyric-to-melody generation.
  arXiv  Detail & Related papers  (2022-08-11T08:44:47Z)
• Self-supervised Context-aware Style Representation for Expressive Speech Synthesis [23.460258571431414]
  We propose a novel framework for learning style representation from plain text in a self-supervised manner. It leverages an emotion lexicon and uses contrastive learning and deep clustering. Our method achieves improved results according to subjective evaluations on both in-domain and out-of-domain test sets in audiobook speech.
  arXiv  Detail & Related papers  (2022-06-25T05:29:48Z)
• Genre-conditioned Acoustic Models for Automatic Lyrics Transcription of Polyphonic Music [73.73045854068384]
  We propose to transcribe the lyrics of polyphonic music using a novel genre-conditioned network. The proposed network adopts pre-trained model parameters, and incorporates the genre adapters between layers to capture different genre peculiarities for lyrics-genre pairs. Our experiments show that the proposed genre-conditioned network outperforms the existing lyrics transcription systems.
  arXiv  Detail & Related papers  (2022-04-07T09:15:46Z)

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.