Universal Audio
No mentions found
This entity hasn't been tracked yet, or Iris is still building its knowledge base.
Related Articles from SNS
Universal Audio Volt 876 USB Audio Interface Review: Pro-Level Polish
In the fall of 2006, I decided emo was out and IDM was in. Fueled by the hope of becoming the next Four Tet or Aphex Twin, I marched into my local Guitar Center and purchased an audio interface to convert my guitar and vocals into ones and zeroes, then mangle them in Ableton Live. When I got home, I plugged a brand-new M-Audio Fast Track Pro into my Windows desktop and immediately hit a brick wall of audio driver configuration hell.
USAD 2.0: Scaling Representation Distillation for Universal Audio Understanding
arXiv:2606.06444v1 Announce Type: cross Abstract: Audio encoders are critical to modern audio applications as large language models (LLMs) increasingly rely on a single encoder for diverse inputs. While self-supervised learning (SSL) has yielded strong domain-specific encoders like speech or music experts, multi-domain approaches like USAD and SPEAR remain limited in coverage and evaluation. Recent studies also suggest supervised encoders align better with audio LLMs.
Show HN: FFmpeg WebCLI – Full FFmpeg in Browser, Offline PWA, No Uploads(WASM)
A browser-based video editor powered by ffmpeg.wasm. No uploads, no servers -- all processing happens locally in your browser using WebAssembly. Live app: https://tejaswigowda.com/ffmpeg-webCLI/ - ✅
Where Rectified Flows Leak: Characterising Membership Signals Along the Interpolation Path
arXiv:2606.07271v1 Announce Type: new Abstract: Understanding what generative models retain from training data remains challenging, with implications for copyright and privacy. Beyond verbatim reproduction, models can encode subtler traces of their training data that never surface in their outputs yet remain exploitable. We study this regime for Rectified Flows, which are increasingly used in deployed generative systems.
<em>The Atlantic</em> Announces Editorial Fellowship Class for 2026–27
The Atlantic is announcing six early-career journalists who have been selected for a yearlong editorial fellowship program: Laney Crawley, Catherine Goodman, Nora Lowe, Jack Rodriquez-Vars, Jacob Smollen, and Katherine Weyback. This is The Atlantic’s first class of fellows since 2020; the six joining next month were selected from a pool of more than 1,300 applicants. During their year in the newsroom, the fellows will be embedded with teams to support The Atlantic’s journalism; sharpen their...
SDSU Wired Its Dorms with 1,300 AI Cameras Without Telling Students
San Diego State University spent more than $1.3 million turning its campus into one of the most heavily watched in the California State University system and the students who study and live there learned the full scope from their own newspaper rather than from the administration. University Police finished installing over 1,300 AI-enabled cameras in 2024, threading them through classroom buildings, bookstores, dining areas, parking structures, gyms and the residence halls where students...
Wavelet as Tokenizer: Preliminary Results on a Shared Wavelet Token Schema for Natural Signals
arXiv:2606.02631v1 Announce Type: cross Abstract: This paper studies whether audio, images, and video can share a common wavelet token schema rather than relying on separate modality-specific latent grids. It introduces a preliminary continuous-token model built around a one-level Haar DWT/IDWT frontend, a shared coefficient-token layout, optional structural metadata, lightweight modality value adapters, and a shared token-wise encoder-decoder trunk. On Speech Commands, EuroSAT RGB, and...
UniAudio-Token: Empowering Semantic Speech Tokenizers with General Audio Perception
arXiv:2605.31521v1 Announce Type: new Abstract: Semantic speech tokenizers have become a widely used interface for Audio-LLMs, owing to their compact single-codebook design and strong linguistic alignment. However, their focus on linguistic abstraction induces acoustic blindness, limiting their applicability beyond speech-centric tasks. We propose UniAudio-Token, a framework that empowers semantic tokenizers with general audio perception without compromising speech ability.
Many more US voters support gay candidates, but only if they look and act 'straight,' study finds
Many more US voters support gay candidates, but only if they look and act 'straight,' study finds Stephanie Baum Scientific Editor Andrew Zinin Lead Editor The period between 2018 and 2022, sometimes referred to as "the rainbow wave," featured an unprecedented increase in LGBTQ candidates elected to office. Pete Buttigieg's rise from mayor of South Bend, Indiana, to U.S. secretary of transportation with a 2020 bid for president in between sparked a national dialogue about whether gay...
Assessing True Generalisability of Audio-Visual Speech Recognisers
Announce Type: cross Abstract: Current Audio-Visual Speech Recognition (AVSR) models achieve near-perfect performance on the standard LRS3 benchmark, raising concerns of adaptive overfitting. To systematically assess true generalisability, we construct a highly controlled, unseen evaluation set subsampled from the massive MultiVSR dataset. Unlike standard out-of-distribution benchmarks, our subset strictly matches the acoustic, visual, and demographic distributions of the LRS3 test set.