Tag
AMBIT是一个统一框架，通过将自回归场景规划与rectified-flow渲染耦合，实现第一阶Ambisonics（FOA）空间音频的生成与编辑，并利用FOA感知的连续VAE与on-policy自蒸馏来保持空间结构和声学保真度。
Noiz AI 与香港科技大学联合发布 Helix World，一个基于单张图像的实时交互式视听世界模型，可在单块 NVIDIA H800 GPU 上以 24fps 同时生成可探索的视频场景与空间立体声，代码、checkpoint 和在线 demo 均已开源。
作者讲述了如何通过黑盒测试与信号分析，逆向工程一款商业空间音频处理程序，并在 Linux 上使用 PipeWire 复现其效果，涵盖了 HRTF、房间响应与心理声学的实践探索。
WorldSonus is an interactive video-to-audio framework that brings real-time spatial sound synthesis to world models, using a streaming causal autoregressive diffusion architecture with a real-time factor of 0.41, chunk-indexed prompt scheduling for mid-stream sound control, and stereo supervision for spatial alignment.
提出StereoBind框架，通过动态空间对应（Dynamic Spatial Correspondence）将视觉源运动与立体声生成绑定，用于沉浸式联合视频-音频生成，并构建了StereoWorld-29K数据集与评测基准。
HelixWorld is a real-time interactive audio-visual world model where visual scenes and camera-grounded spatial stereo sound co-evolve under user interaction, using a teacher-student distillation pipeline to run at 24 FPS on a single GPU. The paper also introduces HelixBench to evaluate spatial-acoustic consistency of synthesized sound fields.
The paper introduces VGBench, a diagnostic benchmark for action-level addressedness in audio LLM voice agents, and VoxGate, a post-training approach that substantially improves muting of commands from non-wearers (side-talk, self-talk, speaker-switch scenarios) while retaining correct tool calls.
Apple is opening a state-of-the-art live music venue called Apple Music Hall in London, featuring advanced audio and production systems, marking its return to live music events.
OmniEcho introduces a spatially aware omni-modal model and a new benchmark for spatial audio-visual perception and navigation in embodied agents, achieving state-of-the-art performance.
flat.spatial is a new spatial 3D online meeting app that provides playful virtual spaces with spatial audio, games, and collaboration features for events, teams, and education.
Bose announced second-generation QuietComfort headphones with improved comfort, adaptive ANC, and TrueSpatial immersive audio with head tracking, priced at $359.
Skullcandy launches the Crusher 1080 ANC headphones, featuring Bose's QuietControl active noise cancellation and TrueSpatial head-tracking audio, with adjustable bass boosting and up to 60 hours of battery life.
SwanSphere proposes a unified streaming framework for high-fidelity spatial audio generation from panoramic videos and text prompts using causal autoregressive diffusion transformers and multimodal learning strategies, achieving superior performance in both video-to-spatial and text-to-spatial audio tasks.
Sennheiser announces Momentum 5 Wireless headphones with improved ANC, a user-replaceable battery, and new codec support including AptX Lossless, priced at $399.99 and available June 30.
Google is experimenting with HP Dimension's immersive display and spatial audio to make Google Beam group meetings feel more inclusive by showing remote participants in true size and anchoring voices to speakers, improving social connection by 50%.
Prostir zvuku is a spatial nature sound mixer for Mac.