StepFun has released StepAudio 3 ASR, ranking #1 on the AA-WER Index for non-streaming Speech to Text with 1.7% WER, a notable improvement on StepAudio 2.5 ASR (4.7%)
StepAudio 3 ASR is StepFun's new Speech to Text model, available through the StepFun API for non-streaming transcription, and the first StepFun model to reach the top of our non-streaming Speech to Text leaderboard.
Key takeaways
➤ Accuracy: StepAudio 3 ASR scores 1.7% on the AA-WER Index, effectively tied with Alibaba's Fun-Realtime-ASR-preview at 1.7%. It leads on AA-AgentTalk, our held-out voice-agent dataset, at 1.4%, but trails on long-form Earnings22 calls at 2.8%, where Fun-Realtime-ASR-preview scores 1.8%
➤ Speed: The model transcribes at a speed factor of 88x real time, behind MAI-Transcribe-2 at 374x, Smallest AI Pulse Pro at 285x and Grok Voice Transcribe 2.0 at 154x, roughly level with ElevenLabs Scribe v2 at 84x and Gemini 3.5 Transcribe at 91x, and ahead of Fun-Realtime-ASR-preview at 20x
➤ Price: StepAudio 3 ASR costs $0.40 per hour, or $6.67 per 1,000 minutes, the most expensive of the five most accurate models. MAI-Transcribe-2 and Grok Voice Transcribe 2.0 cost $1.67 per 1,000 minutes and…
StepFun's StepAudio 3 ASR tops Artificial Analysis's non-streamingSending a model's response token by token as it is generated, so the reader sees text immediately instead of waiting for the whole answer.Full definition → Speech-to-Text leaderboard at 1.7% word error rate, sharply better than its predecessor StepAudio 2.5 ASR's 4.7%.
It's effectively tied overall with Alibaba's Fun-Realtime-ASR-preview (also 1.7% WER), leads on the AA-AgentTalk voice-AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → dataset (1.4%), but trails on long-form Earnings22 calls (2.8% vs 1.8%).
Despite top accuracy, StepAudio 3 ASR transcribes at only 88x real-time speed, well behind MAI-Transcribe-2 (374x), Smallest AI Pulse Pro (285x), and Grok Voice Transcribe 2.0 (154x).
At $0.40/hour ($6.67 per 1,000 minutes), it's the most expensive of the five most accurate models, roughly 4x MAI-Transcribe-2/Grok Voice Transcribe 2.0 and nearly 2x ElevenLabs Scribe v2 ($3.67/1,000 min).
Terms in this piece · Glossary
streaming — Sending a model's response token by token as it is generated, so the reader sees text immediately instead of waiting for the whole answer.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
Read the full post on X ↗
