Reproduce “Hy ASR 3.0 Technical Report” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Aoxiong Yin, Auden Li, Chao Huang, Chuanyang Hu, Eitan Wang, Ethan Zou, Fangming Feng, Guangyu Chen and 55 more
Hy ASR 3.0 combines speech recognition with supplied context and clues within an utterance. It uses an audio encoder and language model decoder, with retrieved hotwords or dialogue context supplied as text. On an internal-context test, it reports 61.88% target-keyword recall versus Seed 2.0 Lite’s 54.64%. This is a cross-system comparison, not a same-model test of context effects.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.