StepAudio 2.5 ASR

transcriptionproprietary

4B-parameter speech recognition model with multi-token prediction. Processes ~5 minutes of audio per second of compute. Supports ITN.

Capabilities

Infrastructure
Streaming

Details

Model IDstepaudio-2.5-asr
Provider StepFun
Creator stepfun
Familystepaudio
Licenseproprietary
Statusactive
Typetranscription
Input modalitiesspeech
Output modalitiestext
Open weightYes
Parameters—
Architecture—
Languages—
Release date—
Sourceofficial

Endpoints

audio transcriptionsPOST
https://api.stepfun.com/audio-transcriptions

Pricing

0.15 元／小时 (CNY)

Family Comparison: stepaudio

ModelLanguages
StepAudio 2 ASR Pro—
StepAudio 2.5 ASR—
StepAudio 2.5 TTS—

API

GET/v1/models/stepfun/stepaudio-2.5-asr