JIS: A Speech Corpus of Japanese Idol Speakers with Various Speaking Styles

2025-06-23

AI Summary

Current speech synthesis research is hindered by the scarcity of high-quality, culturally specific speech corpora—particularly for underrepresented speaker identities such as Japanese young female live idols—leading to inadequate speaker similarity evaluation in TTS and voice conversion (VC) systems and impeding listener-preference-driven personalized voice modeling. To address this, we introduce JIS, the first open-source, ethically compliant speech corpus specifically designed for this demographic. JIS encompasses diverse spoken styles and features fine-grained metadata annotation, speaker anonymization, and contextual cultural documentation. It fills a critical gap in culturally grounded speech resources and significantly improves speaker identifiability modeling accuracy. As a benchmark, JIS enables more rigorous similarity assessment in TTS/VC and facilitates preference-aware voice generation. The corpus is freely available for non-commercial use and accompanied by analytical tools and practical guidelines.

Technology Category

• Natural Language Processing: Speech
• Application Domains: Humanities & Computational Social Science
• Humans and AI: AI for Accessibility

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Systems and Infrastructure for Web, Mobile and WoT: Decentralized Web and Fediverse systems

Abstract

We construct Japanese Idol Speech Corpus (JIS) to advance research in speech generation AI, including text-to-speech synthesis (TTS) and voice conversion (VC). JIS will facilitate more rigorous evaluations of speaker similarity in TTS and VC systems since all speakers in JIS belong to a highly specific category: "young female live idols" in Japan, and each speaker is identified by a stage name, enabling researchers to recruit listeners familiar with these idols for listening experiments. With its unique speaker attributes, JIS will foster compelling research, including generating voices tailored to listener preferences-an area not yet widely studied. JIS will be distributed free of charge to promote research in speech generation AI, with usage restricted to non-commercial, basic research. We describe the construction of JIS, provide an overview of Japanese live idol culture to support effective and ethical use of JIS, and offer a basic analysis to guide application of JIS.

Problem

• Constructing a Japanese Idol Speech Corpus for AI research
• Enhancing speaker similarity evaluation in TTS and VC systems
• Enabling research on listener-specific voice generation preferences

Innovation

• Constructs Japanese Idol Speech Corpus (JIS)
• Facilitates speaker similarity evaluations in TTS/VC
• Enables voice generation tailored to listener preferences