It is a common practice among saxophonists to transcribe songs. The lyrics, an essential element of a song, are not included in songs transcriptions for saxophone and consequently a substantial layer of the composition is lost. This dissertation addresses the problem of song transcriptions by saxophonists and provides strategies for improving music performance through linguistics. In detail, the aim of this research is to study the three following topics: (i) the influence of vowels in shaping the saxophone timbre, (ii) the influence of consonants on the articulation and (iii) the impact of linguistic stress to agogic and musical phrasing. I have developed an experiment, applying a convenience sample procedure to measure the relation and influence of linguistic to musical features. A group of highly trained saxophonists participated in the research experiment to evaluate how linguistics improves musical performance. The results of the experiment demonstrate that vowels and consonants provide a deeper conscience and ability to apply a larger variety of timbre and articulation, in particular in the context of song transcriptions. Furthermore, the results show that the analysis of linguistic stress improves the quality of the musical phrasing of songs’ transcriptions.