The paper adopts comparative and content research methodology to carry out the quality study of Machine Translation (MT) in Translating Chinese Music Historical Texts. The study selects Youdao AI Translate (labeled as MT1), Baidu Translate (labeled as MT2) and Tencent Translate (labeled as MT3) as the machine translation tools for comparative study and adopts a Likert scaling evaluation method to assess the performance of each MT tool. The result shows that the three MT systems are underperformance in translating ancient Chinese music historical texts, with varying degrees of errors primarily in terminology translation, semantic misunderstandings, and mistranslations of culturally specific expressions. In terms of fidelity, the mistranslation of terminology is particularly pronounced, with MT1 showing slightly better results. All three systems exhibit cultural omissions, meaning they convey only the literal meanings while neglecting cultural imagery. In the dimension of fluency, the three ma-chine translation systems handle the syntax of the texts relatively well; however, the main issue arises from misinterpretations of meaning, leading to logical inconsistencies and reading difficulties. Regarding cultural adaptability, all three systems demonstrate significant cultural misreading, primarily stemming from a lack of understanding of the source culture and its contextual references. There are some inspirations for translation teaching based on the research findings which are understanding the limitations of MT, emphasizing the importance of cultural context in translation, integrating post-editing skills and focusing on terminologies and specialized fields.