A quantizer design approach is presented that falls into the switched quantizer class. However the actual technique used is more like the techniques used for fixed quantizer design. The motivation for this approach was the fact that most natural processes can be viewed as the sum of several subprocesses. An example of this is the speech process which can be broken down into voiced speech and unvoiced speech. The problem that arises here is that the switching algorithm needs to be fairly complex if it is to classify an input sample as belonging to either of these processes. The authors take a somewhat simplistic approach towards breaking a process down into subprocesses in order to come up with a simple switching algorithm.