These days, emotional content analysis and AI-based music techniques are crucial for understanding the emotional resonance of lyrics across many linguistic domains. In order to methodically investigate the emotional congruence in English translations of German lyrics, we developed a special dataset for this study. Using advanced natural language processing techniques DistilBERT and NRC+Kmeans, we assessed the emotional coherence between the original and translated text. According to our research, translations have the power to significantly change the original emotional profile of lyrics as well as the emotional intensity and diversity of those lyrics. English translations showed crisper, more distinct, and intense emotional profiles, whereas German songs featured more uniform and intertwined emotional profiles. These results highlight the significant impact that translation has on the subtle emotional undertones of lyrics. Although certain German songs were more upbeat than their English versions, our findings demonstrated that general balance in positive emotions was preserved. It was discovered that when it came to negative emotions, translation softened some of them and made the English translations less negative. In the DistilBERT model, the categorization consistency of emotions was 81.84%, whereas in the NRC EmoLex language, it was 63.11%. The DistilBERT model performed better in song classification, improved emotional consistency across both languages, and was particularly effective at expressing positive emotions.