Show HN: Teach a speech model (en) a name by saying it once, on your Mac
I wanted to see if a speech model could spell a name better if it heard it, not just read it. Most STT tools handle custom words as text (keyterm list, fixing transcript after with llm, etc). So I trained an adapter to try it, specifically a LoRA on Qwen3-ASR 1.7b (see benchmark). This LoRA transcribes words only after it hears a recording of each of your custom words. To make this model easier to use in day-to-day life, we also built Detail Keeper. It's a Mac app where you can put this adapter
阅读原始来源
来源记录
指标快照
points · 2 votes
2026-11-05 21:15:31.861Z
cumulative since publication
研究分析
研究主张
来源描述：I wanted to see if a speech model could spell a name better if it heard it, not just read it. Most STT tools handle custom words as text (keyterm list, fixing transcript after with llm, etc). So I trained an adapter to try it, specifically a LoRA on Qwen3-ASR 1.7b (see benchmark). This LoRA transcribes words only after it hears a recording of each of your custom words. To make this model easier to use in day-to-day life, we also built Detail Keeper. It's a Mac app where you can put this adapter
评测与比较记录
来源票数: 2 votes (2026年11月5日 UTC; cumulative since publication) 可供比较的近期同类记录：“Memorilla: Latent Semantic Memory for LLMs” (Hacker News, 2026年11月6日)；“Show HN: Ru-Jev – an open model that scores caller-supplied options on CPU (MIT)” (Show HN, 2026年11月6日)。主题相近仅用于导航，未核验其是否属于同一需求或独立证据。
结论适用范围
结论只能在所述模型版本、任务、数据与推理预算内解释。社区票数、下载量或榜单位置不能证明另一工作流中的可靠性。
有效性与复现缺口
需核对样本选择、基线配置、不确定性、数据污染风险、负面结果以及代码或权重可获取性；此处没有本站复现实验记录。
复核路线
阅读论文或模型卡及评测附录，针对明确基线复现一个有限结论，再测试代表性用户任务并记录结论失效的场景。