The rapid advancement of AI has transformed education and language learning, leading to the development of Computer Aided Language Learning (CALL) systems. These systems help learners practice reading, writing, pronunciation, and vocabulary. German learners face specific challenges with grammar, pronunciation influenced by native accents, and vocabulary, compounded by classroom constraints like limited time and feedback. Research highlights the importance of feedback alongside assessment for language improvement. While existing CALL systems address pronunciation and grammar correction for various languages, most focus on a single skill. For German, systems are scarce, offering limited feedback and lacking adaptive learning features. To address these gaps, a unified, web-based CALL system for German was developed, integrating pronunciation assessment and grammar correction with dynamic feedback and adaptive learning. The pronunciation system includes two modules. The first uses a Siamese model and Allosaurus for phoneme recognition, achieving 74% accuracy and providing detailed feedback on mismatches and similarities. A Smart Learning Algorithm adapts to learners’ mistakes, improving retention. The second module evaluates any German word or sentence, categorizing feedback as “Needs more practice,” “Good,” or “Excellent” with tailored tips. A transformer-based model generates corrections for grammar correction, while an LLM-based system provides explanations and insights. A thesaurus module enhances vocabulary, and stored data-correction pairs improve future GEC models while protecting personal data. The system was evaluated using real-world data, and it showed promising results. Future work includes enhancing feedback, user testing, and expanding to additional language skills, offering a comprehensive, learner-focused tool for German language mastery.