전자책 TTS 음성 생성ㆍ전송과 배타적발행권 침해에 관한 소고 – 서울고등법원 2025. 6. 19. 선고 2024나 2011618 판결을 중심으로 -

With the expansion of e-books and digital reading platforms, text-to-speech (TTS) technology has emerged as a key competitive feature in platform-based reading services. TTS functions not only as an accessibility tool that substantially enhances reading access for information-vulnerable groups such as persons with visual impairments and older adults, but also as a new mode of content distribution closely intertwined with the rapid growth of the audiobook market. Nevertheless, significant legal uncertainty remains regarding the copyright characterization of converting textual works into audio through TTS and the scope of rights that may be implicated by such practices. These issues came to the forefront in the dispute between KT Millie’s Library and Willa concerning the provision of audiobook and TTS services. The core legal question was whether KT Millie’s Library’s practice of converting e-book text into audio via TTS infringed Willa’s exclusive publication rights relating to audiobook production and distribution. Departing from the lower court’s ruling, the Seoul High Court held that the platform’s TTS service went beyond a merely auxiliary accessibility function and instead constituted a substantive transformation of the mode of exploiting copyrighted works, thereby infringing exclusive publication rights in audio content. The Court emphasized that the audio conversion involved in the TTS process was not limited to facilitating access, but rather amounted to the reproduction and communication of the work through an independent audio medium. Furthermore, noting that the platform planned, designed, and controlled the entire process of generating, deleting, and transmitting audio files (WAV), the Court identified the platform—not the user—as the substantive actor responsible for reproduction and transmission. The Court also interpreted the relevant contractual provisions to mean that, where TTS was explicitly included within the definition of “audio content,” the scope of exclusive publication rights extended to all text-to-audio uses regardless of the specific technical implementation. Against this backdrop, this article analyzes whether the automated generation of audio from e-book text via TTS constitutes a form of reproduction encompassed by exclusive publication rights under copyright law, and how the substantive actor responsible for reproduction and transmission should be identified in platform-based services. It further examines whether copyright limitations may apply in light of the technological and social objectives of accessibility. Finally, the article argues for a more refined legal characterization of TTS and other audio conversion technologies and explores normative criteria for distinguishing accessibility-oriented uses from commercial services, with the aim of proposing interpretive and legislative directions that balance copyright protection, technological innovation, and users’ rights of access.