arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.23387cs.SDcs.AIcs.CLeess.AS
收藏

Hello-Chat: 向真实社交音频交互迈进

Hello-Chat: Towards Realistic Social Audio Interactions

Yueran Hou, Peilei Jia, Zihan Sun, Qihang Lu, Wenbing Yang, Yingming Gao, Ya Li, Jun Gao

AI总结:

Hello-Chat通过大规模真实对话数据和模态交错训练策略,实现了在拟人化生成和情感一致性方面的突破,推动 empathetic AI 的发展。

摘要:

近年来,大型音频语言模型(LALMs)在语音识别和翻译方面展现了卓越的性能。然而,现有模型往往存在感知与表达之间的脱节,导致一种机械的

英文摘要:

Recent advancements in Large Audio Language Models (LALMs) have demonstrated exceptional performance in speech recognition and translation. However, existing models often suffer from a disconnect between perception and expression, resulting in a robotic "read-speech" style that lacks the spontaneity and emotional resonance of real human interaction. In this report, we introduce Hello-Chat, an end-to-end audio language model designed for realistic social scenarios. By leveraging a massive dataset of real-life conversations and employing a modality-interleaved training strategy, Hello-Chat achieves a breakthrough in anthropomorphic generation. Experimental results show that our model not only reaches state-of-the-art (SOTA) performance on specific audio understanding tasks but also significantly outperforms existing baselines in prosodic naturalness and emotional alignment, paving the way for the next generation of empathetic AI agents.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑