Step ASR 1.1
Step ASR 1.1 has separate file and streaming identifiers. The current live-stream guide marks the stream variant as being phased out and recommends StepAudio 2.5 ASR Stream.
Deprecated API; check access details
Treat file and live paths separately
The file guide still names step-asr-1.1 as an example, while the streaming guide has an explicit deprecation label. These statements do not establish a single retirement date for both interfaces.StepFun asynchronous audio file recognitionStepFun bidirectional speech recognition
Review live transcript events when changing models
Practical guidance. A replacement can return partial results and final text differently. Check how the application joins those events, and how it handles revised words. Keep the final transcript linked to the same audio timestamps.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Deprecated API; check access details |
| Access | API |
| Model identifiers | step-asr-1.1step-asr-1.1-stream |
| Deprecation scope | The stream variant is gradually deprecated. The reviewed guide gives no final shutdown date.StepFun bidirectional speech recognition |
| Billing status | Both 1.1 identifiers remain in the billing catalog.StepFun billing catalog and 9 November 2026 audio change |
Related models
Step ASR supports Chinese and English speech recognition.
Launch date not confirmedAPI- 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