Design of Automatic Translation Algorithm for English and American Literature Text Based on Deep Learning

British and American literature, as a precious heritage of Western culture, has always been an important bridge for cultural exchange. However, translating it into other languages, especially those of non English speaking countries, often faces multiple challenges. These challenges not only stem from differences in language structure, but also involve deep interpretation of culture, emotions, and rhetorical techniques. Traditional machine translation methods, due to their inherent limitations, are difficult to perfectly address these challenges. This article explores an automatic translation algorithm for English and American literary texts based on DL (Deep Learning) to address this issue. Finally, the research results indicate that regardless of the number of statements being 1100, 1500, 2100, or 3300, the translation quality evaluation accuracy of our algorithm is much higher than the other two algorithms. The highest accuracy of our algorithm can reach 95%, and it is more stable and has certain application value. Overall, this study provides a new and efficient method for automatic translation of English and American literary texts, which not only advances in technology but also opens up new perspectives and possibilities for future cross-cultural communication and research.