Speech synthesis is crucial for human-computer interaction, enabling natural and intuitive communication. However, existing datasets involve high construction costs due to manual annotation and suffer from limited character diversity, contextual scenarios, and emotional expressiveness. To address these issues, we propose DialogueAgents, a novel hybrid agent-based speech synthesis framework, which integrates three specialized agents -- a script writer, a speech synthesizer, and a dialogue critic -- to collaboratively generate dialogues. Grounded in a diverse character pool, the framework iteratively refines dialogue scripts and synthesizes speech based on speech review, boosting emotional expressiveness and paralinguistic features of the synthesized dialogues. Using DialogueAgent, we contribute MultiTalk, a bilingual, multi-party, multi-turn speech dialogue dataset covering diverse topics. Extensive experiments demonstrate the effectiveness of our framework and the high quality of the MultiTalk dataset. We release the dataset and codethis https URLto facilitate future research on advanced speech synthesis models and customized data generation.
View on arXiv@article{li2025_2504.14482,
title={ DialogueAgents: A Hybrid Agent-Based Speech Synthesis Framework for Multi-Party Dialogue },
author={ Xiang Li and Duyi Pan and Hongru Xiao and Jiale Han and Jing Tang and Jiabao Ma and Wei Wang and Bo Cheng },
journal={arXiv preprint arXiv:2504.14482},
year={ 2025 }
}