This　paper　presents　a　new　algorithm　to　identify　matrix　A　is　an　element　of　R-mxn　knowing　only　their　multiplication　X　=　AS.　Where　S　is　an　element　of　R-nxN　is　sparse　and　M　＜　n.　The　data　used　for　matrix　identification　are　chosen　by　Least　Square　method,　whose　fitting　errors　are　smaller　than　a　given　threshold.　Then,　K-means　clustering　method　is　adopted.　This　technique　avoids　data　overlapping　at　the　origin,　thus　improving　the　accuracy　of　mixing　matrix　estimation.　The　validity　of　the　method　for　true　voice　separation　is　verified　by　computer　simulation.　Also　comparison　with　other　methods　is　made　to　verify　the　efficiency　of　the　algorithm.　Simulations　show　that　the　algorithm　has　the　property　of　accuracy　and　low-cost　computation.