In　the　milling　process,　the　stability　lobe　diagrams　(SLD)　allow　the　selection　of　appropriate　milling　parameters　to　maximize　the　machining　efficiency.　While　calculating　the　SLD　is　time-consuming,　this　paper　proposes　a　method　for　predicting　the　SLD　based　on　spindle　speed　and　depth　of　cut　using　deep　learning.　The　prediction　model　constructed　by　this　algorithm　using　LSTM　neural　network　can　predict　the　stability　boundary　according　to　different　cutting　depths　and　spindle　speeds.　Validation　is　carried　out　through　experiments,　and　the　results　show　that　the　proposed　method　has　a　stability　prediction　accuracy　of　96.3%　under　different　operating　conditions.　It　is　also　verified　that　the　proposed　method　can　reasonably　select　milling　parameters　and　improve　machining　efficiency.　©　2024　SPIE.