Esophageal speech differs from laryngeal speech primarily with regard to the sound or source of voicing. Esophageal speech is perhaps the ideal method, since it is achieved using the potential of the patient’s existing structure. Even so, it should be the case that esophageal speakers have difficulty in accomplishing the voiceless distinction for consonants. Surprisingly, perceptual studies often reveal that there is a clear voicing distinction resulting good articulation scores in skilled esophageal speech. The purpose of the present study was to relate the three-way distinction of Korean voiceless stops in manner of articulation with normal speakers and skilled esophageal speakers in terms of the voicing distinction in consonants. Perceptual study was first performed to examine the pattern of the confusion that normal listeners would make in comprehending initial consonants spoken by esophageal speakers. Finally, acoustic analysis were performed to investigate the acoustic characteristics of esophageal speech compared to the normal speech with special reference to the voiceless distinction.