S-VoCAL: A Dataset and Evaluation Framework for Inferring Speaking Voice Character Attributes in Literature

2026-03-01

AI Summary

This work addresses the lack of effective methods and evaluation benchmarks for inferring vocal attributes—such as age, gender, and accent—of fictional characters in synthetic audiobooks. To bridge this gap, the authors introduce the first benchmark specifically designed for character-oriented voice attribute inference in literary texts, encompassing eight sociophonetic attributes across 952 character–book pairs derived from Project Gutenberg. The approach leverages large language model embeddings with similarity-based metrics and retrieval-augmented generation (RAG) techniques. Experimental results demonstrate strong performance on attributes like age and gender, while highlighting ongoing challenges in more nuanced dimensions such as accent and health status. This study provides the community with the first standardized dataset, evaluation framework, and baseline results for voice attribute inference in narrative fiction.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

With recent advances in Text-to-Speech (TTS) systems, synthetic audiobook narration has seen increased interest, reaching unprecedented levels of naturalness. However, larger gaps remain in synthetic narration systems' ability to impersonate fictional characters, and convey complex emotions or prosody. A promising direction to enhance character identification is the assignment of plausible voices to each fictional characters in a book. This step typically requires complex inference of attributes in book-length contexts, such as a character's age, gender, origin or physical health, which in turns requires dedicated benchmark datasets to evaluate extraction systems' performances. We present S-VoCAL (Speaking Voice Character Attributes in Literature), the first dataset and evaluation framework dedicated to evaluate the inference of voice-related fictional character attributes. S-VoCAL entails 8 attributes grounded in sociophonetic studies, and 952 character-book pairs derived from Project Gutenberg. Its evaluation framework addresses the particularities of each attribute, and includes a novel similarity metric based on recent Large Language Models embeddings. We demonstrate the applicability of S-VoCAL by applying a simple Retrieval-Augmented Generation (RAG) pipeline to the task of inferring character attributes. Our results suggest that the RAG pipeline reliably infers attributes such as Age or Gender, but struggles on others such as Origin or Physical Health. The dataset and evaluation code are available at https://github.com/AbigailBerthe/S-VoCAL .

Problem

• voice character attributes
• fictional characters
• attribute inference
• synthetic narration
• sociophonetics

Innovation

• S-VoCAL
• character voice attributes
• sociophonetics
• LLM embeddings
• Retrieval-Augmented Generation