Pronunciation is a crucial component of English language learning that significantly influences learners' speaking performance and communicative competence. However, many vocational high school students still experience difficulties in producing accurate pronunciation due to limited practice opportunities, lack of motivation, and conventional teaching methods. In response to this issue, the integration of technology, particularly Automatic Speech Recognition (ASR), offers potential support for improving pronunciation learning by providing immediate feedback and opportunities for independent practice. Therefore, this study aims to explore the use of ASR in learning pronunciation skills and to investigate students' perceptions toward its use among eleventh-grade students of SMK Pembangunan Bukittinggi. This study employed a qualitative research design using a qualitative survey approach. The participants consisted of 15 eleventh-grade students selected through purposive sampling, along with one English teacher as a supporting informant. Data were collected through classroom observation and semi-structured interviews focusing on students' experiences and perceptions of using ASR in pronunciation learning. The data were analyzed using the interactive model proposed by Miles and Huberman, which includes data reduction, data display, and conclusion drawing. The findings of the study indicate that ASR, particularly Google Gemini and ChatGPT, was used by students to support pronunciation practice and receive real-time feedback. The results also reveal that students generally had positive perceptions toward the use of ASR, as it helped them improve pronunciation accuracy, increased learning motivation, and reduced speaking anxiety.