RICHARD ROBLES retweetledi

StepAudio 2.5 ASR is here!
4x faster decoding. 30-minute continuous audio. $0.022/hour.
We built this to push ASR further — faster inference, longer inputs, bilingual accuracy that holds up, and a price point that actually scales.
⚡ 4x decoding speed vs. previous generation
⚡ 30-min continuous audio processing
⚡ Top-tier bilingual (EN/ZH) performance
⚡ 80% lower inference cost — API at $0.022/hr
For more details, check out:
- Documentation: platform.stepfun.ai/docs/en/guides…
- Demos: stepaudiollm.github.io/step-audio-2.5…
- Model card: stepaudiollm.github.io/step-audio-2.5…
English



















