Tagged articles

Audio Modeling

2 articles · Page 1 of 1
Weekly Large Model Application
Weekly Large Model Application
Aug 10, 2026 · Artificial Intelligence

LiveKit Turn Detector v1.0 Beats Deepgram Flux and Makes Turn Detection Measurable

LiveKit’s Turn Detector v1.0 replaces the traditional text‑only pipeline with a dual‑branch audio‑semantic architecture, achieving a 9.9% error rate at 300 ms latency—outperforming Deepgram Flux’s 12.9%—and releases an open‑source benchmark (eot‑bench) that turns end‑of‑turn detection into a reproducible engineering problem.

Audio ModelingDeepgram FluxLLM Fusion
0 likes · 11 min read
LiveKit Turn Detector v1.0 Beats Deepgram Flux and Makes Turn Detection Measurable
Weekly Large Model Application
Weekly Large Model Application
Mar 22, 2026 · Artificial Intelligence

Inside MiMo-Audio: Dissecting the Large-Scale Audio Model

The article breaks down MiMo-Audio, a next‑token‑prediction‑style large‑scale audio model built on Qwen2, detailing its acoustic front‑end, RVQ tokenizer, patch‑based transformer architecture, streaming capabilities, performance advantages, engineering constraints, and recommended application scenarios.

Audio ModelingFew-shotQwen2
0 likes · 9 min read
Inside MiMo-Audio: Dissecting the Large-Scale Audio Model