StepAudio Voice AI: Real-Time Speech Synthesis
StepFun has released StepAudio 2.5 Realtime, a voice model with fully customizable personas. It understands paralinguistics (intonation, emotion), supports Chinese and English, and connects via WebSocket. In April tests, it took first place across all parameters.
AI-processed from MarkTechPost; edited by Hamidun News
StepAudio 2.5 Realtime has been released: an integrated voice system that combines speech recognition and synthesis, with support for customizable personas for roleplay.
What is StepAudio?
It is an integrated voice system that combines speech recognition and synthesis. It connects through a WebSocket API and runs in real time with minimal latency.
Which languages does StepAudio support?
The model supports Chinese and English.
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.