The purpose of this study is to investigate seventh-grade students' opinions about the role of songs in improving speaking skills. The phenomenological design, one of the qualitative research designs, was used in the study. Participants were specified using the maximum variation sampling method, one of the purposeful sampling methods. As the study sample, 7 schools of varied sizes, with different socioeconomic levels, and different opportunities were selected so that every part of Ödemiş district of Izmir province could be represented. The study participants consisted of 26 students studying at these schools. Research data were collected using a semi-structured interview form developed by the researcher. The descriptive analysis method was used in data analysis. In the analysis, firstly, the audio recordings were transcribed. Afterward, these recordings were analyzed. As a result of the interviews, students' opinions about the songs they listened to and their speaking skills were gathered under 3 (three) main themes: "Vocabulary Learning," "Speech Errors," and "Effective Speech." These main themes were divided into various sub-themes, and the sub-themes were divided into certain categories in line with different opinions received from students. Vocabulary learning included 4 (four) sub-themes: "Turkish vocabulary learning," "permanent vocabulary learning," "vocabulary learning only from foreign songs," and "no vocabulary learning." The theme of speech errors comprised 5 (five) sub-themes: "use of a local dialect," "pronunciation errors in songs," "singer's effect," "using foreign words," and "using slang words." The theme of effective speech consisted of 4 (four) sub-themes: "stress, pause, and intonation in songs," "songs and fluent speech," "songs and speech rate," "use of songs in activities."

Speaking skill, song, qualitative reseach, phenomenology.