We propose a system for automatically determining boundaries between phonetic segments in a speech wave given a phonetic transcription: automatic segmentation. The system uses edge detectors that are applied to various speech representations; both are optimized for each diphone or diphone class. Output from these detectors, which contains spuriously detected edges, is then combined with alternative pronunciations generated via rules from the canonical pronunciation. The final output is generated with lowest-cost path algorithms applied to finite state transducers.