Artificial intelligence in audiobooks: applications and perspectives

Abstract

The use of Artificial intelligence (AI) techniques in the context of audiobooks has expanded the possibilities for accessibility, personalization and immersion, covering aspects from voice recognition and synthesis to interactive multimodal experiences and personalized recommendations, in addition to enhancing content retrieval and expanding access to information. This study aimed to identify studies on the use of AI in audiobooks in the academic literature. To this end, a literature review was conducted in the Scopus, Web of Science, ACM Digital Library, IEEE Xplore and Scielo databases, between May and August 2025, resulting in the selection and analysis of 35 articles. The results reveal that the studies focus on four categories: (i) speech recognition; (ii) voice synthesis; and personalization; (iii) voice-based experiences; and (iv) generative AI and LLMs. It was observed that technical studies focused on Automatic Speech Recognition and Voice Synthesis predominate, while voice-based experiences and LLM applications are still emerging, indicating future trends. Audiobooks are also frequently used as technical corpora for model development, with few studies focused on directly improving the user experience, in addition to a scarcity of research in the field of Information Science. It can be concluded that, despite recent advances, there are gaps related to the lack of user-centered studies, the predominant use of audiobooks as a technical corpus as well as few ethical and social aspects. This overview provides theoretical and practical support for future research in the area.

Ähnliche Arbeiten