This study aimed to determine the pronunciation ability of lexicons containing diphthongs and monophthongs in children. Therefore, this study aims to describe the vocabulary skills containing diphthong and monophthong elements in four-year-old children. The subjects of this study were Bintang and Gerald, who were four years and four months old. The data in this study were lexicons containing diphthongs and monophthongs spoken by Bintang and Gerald. This qualitative study uses a cross-sectional design, where the subject is usually more than one person, and the topic has been predetermined. The data was collected using observation methods supported by recording and note-taking techniques. After the data were collected, the data were analyzed based on the problems in this study. The analysis method used was the Miles & Huberman interactive analysis model, which included data reduction, presentation, and conclusions or verification. The study results showed that the research subjects could pronounce monophthong lexicons well and correctly, but there were still pronunciation errors in diphthong lexicons. Bintang was able to pronounce 12 diphthong lexicons and 24 monophthong lexicons. Meanwhile, Gerald could pronounce 10 diphthong lexicons and 28 monophthong lexicons. The pronunciation errors found are: 1) Pronouncing half a sound or not pronouncing one phoneme in a diphthong, and 2) Changing a diphthong into a monophthong sound (monophthongization).