Human voice encodes both identity and paralinguistic cues, yet encoders in large audio-language models (LALMs) rarely balance both aspects. In this work, we present a study toward building a general-purpose voice encoder that captures nuanced voice cues. Through a comprehensive evaluation, we find that multi-task training yields the most balanced representations, whereas contrastive language-audio pretraining (CLAP) primarily improves retrieval without enhancing paralinguistic understanding. Our final encoder, Auden-Voice, also demonstrates strong performance when integrated with LLMs. The code and training recipes will be released with the audio understanding toolkit Auden.
@article{arxiv.2511.15145,
title = {Auden-Voice: General-Purpose Voice Encoder for Speech and Language Understanding},
author = {Mingyue Huo and Wei-Cheng Tseng and Yiwen Shao and Hao Zhang and Dong Yu},
journal= {arXiv preprint arXiv:2511.15145},
year = {2025}
}