Detecting and diagnosing mispronunciations are crucial, for computer-based pronunciation training systems assisting language learners in enhancing their speaking skills. This study presents a method for identifying mispronunciations by incorporating knowledge of the learners language and uncertainty modeling to tackle the challenges posed by diverse languages and speaker backgrounds. The proposed approach combines existing techniques for native language awareness and uncertainty modeling to create a method for detecting mispronunciations. Although no experimental results are included in this paper the framework establishes a basis for research, in this field. The goal of this work is to support the development of computer-based pronunciation training systems and promote language learning.