StepAudio 2 ASR Pro
StepAudio 2 ASR Pro is the 32B recognition model in StepFun's audio catalog. It remains a named option in the current SSE transcription API.
Compare the current transcription options
The endpoint supports this model, StepAudio 2.5 ASR and StepAudio 3 ASR Max. Its current documentation recommends the Max model.StepFun speech recognition with SSE
Compare accuracy on the same audio
Practical guidance. A larger parameter count is not a complete selection rule for transcription. Compare names, code-switching, background noise and final text timing on recordings that match your use case.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | stepaudio-2-asr-pro |
| Parameters | 32B, as listed in the public model catalog.StepFun public model catalog |
| Request path | POST /v1/audio/asr/sse.StepFun speech recognition with SSE |
Related models
- StepAudio 2.5 ASR
StepFun · StepAudio 2.5
StepAudio 2.5 ASR is a 4B speech recognition model that uses multi-token prediction.
Launch date not confirmedAPI - StepAudio 3 ASR Max
StepFun · StepAudio 3
A speech-recognition model combining acoustic modeling with language-model context for Chinese, English and mixed-language recordings.
Launch date not confirmedAPI · Provider product