Integrating Global Englishes into mainstream EFL classrooms remains challenging, particularly in Japan, where textbooks and listening materials still overwhelmingly privilege native-speaker varieties. For educators, this barrier prevents student exposure to audio resources representing diverse English users in a pedagogically appropriate way. This presentation introduces a classroom-based approach using text-to-speech (TTS) technology to address this issue. Drawing on classroom practice at a Japanese university, the presenter demonstrates how Amazon Polly and Narakeet were used to create customised listening materials featuring a range of Global Englishes accents aligned with curricular and course assessment requirements. The TTS-generated audio was embedded into textbook listening activities and supported through scaffolding techniques such as guided noticing tasks and pre-listening discussion. This approach allowed lower proficiency students to engage with English varieties without increasing cognitive load. The presentation will feature example materials, classroom tasks, and student responses, highlighting how TTS can support exposure to English diversity while remaining practical and time-efficient for teachers. Participants will leave with ideas on how to use TTS tools to create listening materials in their own teaching contexts. The session concludes with a discussion of the pedagogical benefits, current limitations, and ethical considerations of using synthetic voices for Global Englishes input.