Brief

last 24h

[2/2] 221 sources

Multi-source AI news clustered, deduplicated, and scored 0–100 across authority, cluster strength, headline signal, and time decay.

TOOL · Mastodon — fosstodon.org English(EN) · 4h

https:// winbuzzer.com/2026/05/25/stepf un-launches-stepaudio-25-for-live-voice-ai-xcxwbn/ Chinese AI lab StepFun has launched StepAudio 2.5 Realtime, an end-to

Chinese AI lab StepFun has released StepAudio 2.5 Realtime, a new end-to-end live voice model. This system handles both audio input and output, incorporating features for persona control and paralinguistic understanding. AI

IMPACT This new voice AI model could enhance real-time audio applications with advanced persona and paralinguistic controls.
- StepFun
- StepAudio 2.5 Realtime
TOOL · MarkTechPost English(EN) · 1d

StepFun Releases StepAudio 2.5 Realtime: An End-to-End Voice Model with Roleplay-Specific RLHF and Paralinguistic Comprehension

StepFun has released StepAudio 2.5 Realtime, an end-to-end speech large language model capable of real-time, customizable persona interactions. The model integrates speech understanding and generation, utilizing a million-scale persona data augmentation and roleplay-specific Reinforcement Learning from Human Feedback (RLHF) to maintain character consistency. A key differentiator is its paralinguistic comprehension, allowing it to perceive user mood and intentions from vocal cues like tone and speech rate, achieving a score of 82.18 on a relevant benchmark. AI

IMPACT Enhances real-time conversational AI with improved persona consistency and paralinguistic understanding.