CapSpeech: Enabling Downstream Applications in Style-Captioned Text-to-Speech
Abstract
Recent advancements in generative artificial intelligence have significantly transformed the field of style-captioned text-to-speech synthesis (CapTTS). However, adapting CapTTS to real-world applications remains challenging due to the lack of standardized, comprehensive datasets and limited research on downstream tasks built upon CapTTS. To address these gaps, we introduce CapSpeech, a new benchmark designed for a series of CapTTS-related tasks, including style-captioned text-to-speech synthesis with sound events (CapTTS-SE), accent-captioned TTS (AccCapTTS), emotion-captioned TTS (EmoCapTTS), and text-to-speech synthesis for chat agent (AgentTTS). CapSpeech comprises over 10 million machine-annotated audio-caption pairs and nearly 0.36 million human-annotated audio-caption pairs. In addition, we introduce two new datasets collected and recorded by a professional voice actor and experienced audio engineers, specifically for the AgentTTS and CapTTS-SE tasks. Alongside the datasets, we conduct comprehensive experiments using both autoregressive and non-autoregressive models on CapSpeech. Our results demonstrate high-fidelity and highly intelligible speech synthesis across a diverse range of speaking styles. To the best of our knowledge, CapSpeech is the largest available dataset offering comprehensive annotations for CapTTS-related tasks. The experiments and findings further provide valuable insights into the challenges of developing CapTTS systems.
Cite
@article{arxiv.2506.02863,
title = {CapSpeech: Enabling Downstream Applications in Style-Captioned Text-to-Speech},
author = {Helin Wang and Jiarui Hai and Dading Chong and Karan Thakkar and Tiantian Feng and Dongchao Yang and Junhyeok Lee and Thomas Thebaud and Laureano Moro Velazquez and Jesus Villalba and Zengyi Qin and Shrikanth Narayanan and Mounya Elhiali and Najim Dehak},
journal= {arXiv preprint arXiv:2506.02863},
year = {2025}
}