audio

Sonic Pi v5, the live coding music environment, has been released with new features and improvements.

Unsloth releases GGUF quantizations of MiniMax-H3, an omni-modal generative system for video with native stereo audio, enabling local execution via sd-cli and other platforms.

The author wrote a pure-MLX runtime for the vision and audio towers of Nvidia's Nemotron Omni model, enabling the full multimodal model to run locally on Apple Silicon Macs. All 23 tests pass against the PyTorch reference, and it achieves 67-152 tok/s on an M5 Max.

This paper introduces KVAE, a family of tokenizers for audio, image, and video designed for text-conditioned generative models, claiming competitive or superior reconstruction and generation quality compared to existing open-source tokenizers. The code and training details are publicly released.

VibeVoice 1.5B runs locally on an iPhone with ~2.2 GB memory and up to 1.28× real-time speed; the author plans to release the xcframework and code for audio.cpp.

Nothing's budget sub-brand CMF has launched its first open earbuds, the CMF Clip Pro, priced at $99 with open-ear design, multi-point Bluetooth, and up to 33 hours of total battery life.

FFmpeg 9.0 is a major release of the popular open-source multimedia framework, offering libraries and command-line tools for audio, video, subtitle and metadata processing.

OpenAI announces that audio now uses a dedicated fast path while reasoning and tool use run asynchronously, and voice-session startup has been reduced from six network round trips to one.

A review of the Nothing Ear (3a) wireless earbuds, highlighting their build quality, fit, and unique Audio Snapshot feature with AI transcription.

Vinyl for Mac turns your Mac into a spinning vinyl player, offering a fun way to enjoy music with a retro aesthetic.

Termixer is a terminal-based DJ mixer written in Rust that integrates with TidalCycles, MPV, and SuperCollider for live audio mixing with EQ, crossfade, and sample pads.

A review of the Daisy One headphones, praising their tactile volume dial and comfort but noting mediocre noise canceling and sound quality for the price.

Wired reviews and recommends the best smartphones that still include a headphone jack in 2026, highlighting models like the Motorola Moto G Stylus and TCL 60 XE Nxtpaper, while discussing the port's decline since Apple removed it.

MiniMax launches H3, an open multimodal generation model that handles text, images, video, and audio, generating up to 15 seconds of 2K video with native stereo sound, and plans to open-source the weights.

The Nothing Ear 3A earbuds offer solid ANC and audio performance at $99, but suffer from occasional bugs and bass-heavy default sound that requires EQ adjustment.

The EPOS H3 Hybrid gaming headset, combining wired and Bluetooth audio, is on sale for $24.99, 75% off its original price.

FreqWave EQ is a browser-based real-time equalizer that lets users customize web audio.

A technical blog post detailing fast integer-based oscillator implementations for the GBA and rp2040, covering saw, square, triangle, and pulse waves, with focus on speed over aliasing reduction.

Release 0.4 of audio.cpp adds C++/GGML inference for Higgs Audio v3 TTS 4B (10x real-time) and Fish Audio S2 Pro, with full GGUF loading and Q8 speed/VRAM gains.

Liso lets you highlight text on any webpage and converts it into audio for listening.