Attack-free WavLM detector ranks audio deepfakes but fails under shift
The best variant scores 8.45% EER on ASVspoof 2019 LA, then 22.12% EER on the 2021 DF shift.
Stories per day · 30 days
1 RSS
Relatedopen-weights2watermarking2anomaly-detection1asvspoof1backward-compatibility1claude-code1collusion1deepfake-detection1detector1document-parsing1ducks1embedder1false-positives1flow-matching1game1generative-ai1generative-models1huggingface1
19 stories, newest first · one card per story
Stories tagged #audio
Video Rebirth publishes HyperFlow H3 Flash, a compressed MiniMax-H3
The 13.7B–13.9B model makes 5.2–15 second video with stereo audio in 4–8 steps, under a restricted MiniMax license.
Averaging two audio copies breaks traceability in personalized watermarking systems
Averaging only two personalized audio copies produces tracing-failure rates of 87.7-99.7% while largely preserving audio quality.
Youtu-Parsing-Omni parses documents, audio, and video
A single 5B checkpoint is reported at 96.96 on OmniDocBench v1.6 and 75.08 average on OmniParsingBench.
Backward Compatibility in Zero-Bit Neural Audio Watermarking
Legacy-chain training plus angular diversity loss makes neural audio watermarks backward compatible.
FLOMSS applies flow matching to four-source music separation
The model separates vocals, bass, drums, and other in full-band stereo and is competitive on MUSDB18-HQ in objective and listening tests.
Learning Where to Split: Dynamic Band Splitting for Audio Source Separation
Dynamic Band Splitting adapts audio frequency bands, gaining 29% relative SI-SDR on DnR v2.
Speech event detector: Timestamped detection of non-verbal vocalizations and background sounds for rich speech transcription
A 5.8M model timestamps 54 non-speech sounds and merges them into Whisper transcripts.
Microsoft: Windows KB5124010 update crashes some games and apps
Microsoft says Windows 11 preview update KB5124010 crashes some games and apps that use AC-3 audio.
3
SheetSage2: Coherent Lead-Sheet Transcription with Synthetic Supervision
SheetSage2 transcribes coherent lead sheets using synthetic MIDI supervision and beats prior systems on most reported metrics.
Revisiting Input Time-frequency Representations in Multi-pitch Estimation for Vocal Ensembles
arXiv paper shows linear STFT inputs beat harmonic constant-Q transform for vocal-ensemble multi-pitch estimation while cutting feature-extraction cost.
VoxMem: Benchmarking Multimodal Memory in Large Audio Language Models
VoxMem finds no large audio language model exceeds 40% memory accuracy at 32K across 177 hours of speech.
AEGIS: Audio Endogenous Guarding via Internal Signals Against Large Audio-Language Model Jailbreaks
AEGIS uses mid-layer risk signals to cut audio-model jailbreak unsafe rates from 17.9% to 0.4%.
Gemini 3.8 text-to-speech says hello
Google released Gemini 3.8 Flash TTS and Flash-Lite TTS for expressive multilingual speech generation.
TeleAntiFraud 2.0: A Refreshable, Profile-Grounded, and Audio-Based Benchmark for Telecom Fraud Detection
TeleAntiFraud 2.0 introduces a monthly refreshed audio benchmark with 900 Chinese calls for realistic telecom fraud detection evaluation.
FRAUDSkill: Structured Frozen-Weight Skill Optimization for Audio Anti-Fraud Detection
FRAUDSkill adapts frozen audio-language models for telecom fraud detection via external skill programs, reaching 73.50% Macro-F1.
Mothersuperior/yue2-mothersuperior-realaudio-tokenizer-v4 — new model trending #30 on Hugging Face
Community release ships a real-audio semantic tokenizer and NAR LoRA for YuE2-3B, enabling artist-style song generation and covers from user recordings.
Tri-PvP: Exposing Modality Bias in Omni-Modal Large Language Models through Perceptual-Propositional Evidence Conflicts
Tri-PvP is a new benchmark revealing robust modality bias in omni-modal LLMs, with vision favoring perceptual signals and audio favoring propositional ones.
What Did I Just Say? Self-Listening for Full-Duplex Speech Models
Researchers propose Self-Listening, a full-duplex speech approach feeding realized model speech back as input to improve interruption recovery.