Taan发布可控音乐模型
摘要
Taan.ai入选YC 2026秋季批次,基础音乐模型支持哼唱、MIDI、图片、视频、脚本输入,可按时间点生成音乐。在250条时间控制提示测试中,指令成功率为80.1%,高于ElevenLabs的28.9%,定时事件误差在±500ms内的占73.2%。
原文片段
Taan AI: The most controllable foundation music model Foundation music models that do what you tell them to, when you tell them to. **TL;DR:** Taan is a foundation music model built for control and multimodal inputs. Hum a melody, name exact notes, or give it an image, a MIDI file, a video, or a sc…