SAPI allows you to create applications without knowing which speech engine will be implemented on user machines. Speech engines distributed by vendors are compatible with all speech applications through SAPI. This gives users the freedom to choose speech engines that meet their needs.
The scenario is analogous to GDI, which allows programs to draw graphics without knowing the kind of display card or monitor the user has.
The Microsoft Speech API (SAPI) provides a way for applications to include text-to-speech and speech recognition technologies with speech engine independence. Text-to-Speech can audibly notify users when events take place. Speech recognition allows users to vocally command a computer to perform tasks such as opening or closing documents, or send e-mail messages.