stepaudio-2.5-realtime

Realtime
Audio
Voice

Standard & Paid Price

Standard Price
MeterStandard Price
Input price$1.5000 / 1M tokens
Completion price$10.0000 / 1M tokens
Cache read price$0.3000 / 1M tokens
Audio input price$1.5000 / 1M tokens
Audio completion price$10.0000 / 1M tokens
Billing details
Vendor list price comparison

Description

StepFun StepAudio 2.5 Realtime — end-to-end speech-to-speech over a WebSocket session at GET /v1/realtime: streaming audio or text in, streaming audio with a text transcript out, with server-side voice activity detection and interruption. Audio is PCM16, 24 kHz, mono; the vendor lists English as the supported language. Set the voice explicitly in session.update: only StepFun's listed voices are accepted.

API endpoints

  • GET /v1/realtime (realtime)