Presentation Empowering Listening Input with AI-Generated Audio: Exploring TTS Applications for EFL Classrooms more

Input is foundational to second language acquisition. With the explosion of digital media, practically unlimited resources are now available to both learners and educators. Yet in practice, while a rich ecosystem of graded readers and skill-focused textbooks support abundant reading input, resources for listening are comparatively scarce, with materials often tied to reading passages, test prep, or authentic content that may not suit a given class. This presents a challenge for educators seeking listening materials with appropriate proficiency, length, content, and language focus. To this end, recent advancements in text-to-speech technology, particularly the improved naturalness, customizable tones, and multi-speaker support, have revolutionized the ease with which educators can create highly tailored listening materials. This presentation will explore the practical integration of AI-generated audio in EFL classrooms, ranging from extensive listening to focused classroom tasks. The presenter will share his own successes and challenges with listening materials such as example dialogues using target vocabulary for fluency practice, alternative versions of reading passages for comprehension tasks, and audio prompts for test items. Survey results from first- and third/fourth-year Japanese university students on the perceived usefulness, engagement, and preferred applications of AI-audio listening tasks will also be presented.