Model Releases

StepAudio real-time voice released, able to understand your tone and subtext

Original title:StepAudio 2.5 Realtime is live! Real-time voice that picks up what you actually mean — tone, pace,...

AI summary

StepAudio 2.5 Realtime is a real-time voice interaction model. Its core advantage is perceiving users' paralinguistic features, such as tone, rhythm, pauses, and even slight sighs, to understand the true intent behind speech. The model supports highly customizable character personas and speaking styles via API, includes more than 10,000 composable preset characters, and offers five out-of-the-box preset characters for users to try. It is optimized with RLHF and can stably maintain its persona in complex role-playing stress tests. It supports bilingual Chinese-English interaction.

Key Takeaways

  • StepAudio 2.
  • 5 Realtime is a real-time voice interaction model.
  • Its core advantage is perceiving users' paralinguistic features,such as tone,rhythm,pauses,and…
  • The model supports highly customizable character personas and speaking styles via API,includes…
  • It is optimized with RLHF and can stably maintain its persona in complex role-playing stress tests.

Quick Facts

Published
Updated
Source
X:阶跃星辰 StepFun (@StepFun_ai)
Category
Model Releases
AI score
76
Read the original Read the full article at the original source
Why it matters

StepFun 这个语音模型把副语言感知做进了实时 API,产品人想搞点情感化语音交互的可以试试,预设角色能省不少调 prompt 的功夫。

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: StepAudio real-time voice released, able to understand your tone and subtext. May 24, 2026. https://new.nat6.net/en/news/cmpixygw410zxsljwqtlbvr66/

View all →