The rapid globalization of regions with different languages necessitates more advanced non-native tongue proficiency among individuals who need to communicate across language barriers. To deal with this demand, the research conducts a comparative analysis of the acoustic features of similar sounds between standard Mandarin (L2) and Mandarin spoken by Uyghurs (L1). Flege’s Speech Learning Model as well as Lado’s Contrastive Analysis Hypothesis serve as the foundation for this study. First, an acoustic analysis of Mandarin Chinese vowels produced via Uyghur and native Mandarin speakers is carried out. Ten Mandarin speakers and ten Uyghur Chinese speakers were compared with respect to word-level productions, and the 1st, 2nd formant frequencies (F1, F2) for Mandarin vowels were studied. The comparative acoustic analysis results show that the production of four ‘similar’ vowels by male and female Uyghur Chinese speakers differs significantly from native Mandarin speakers. Second, the following features are seen when native Mandarin speakers and Uyghur Chinese speakers compare their vowel spaces: Uyghur speakers’ vowels have a lower F1, which means that their overall vowel space distribution is surpassing that of Chinese speakers’. Thirdly, the Euclidean Distance analysis of monophthongs produced via Uyghur as well as Mandarin Chinese speakers reveals a gender-specific applicability of the Speech Learning Model. While the model successfully explains the production of Mandarin monophthongs by Uyghur female speakers, it falls short in accounting for the production patterns observed in Uyghur male speakers. The research findings of this thesis contribute a new direction to future developments in research aimed at improving speech recognition and Computer-Assisted Language Learning (CALL) systems.