StepAudio 3 series speech models free on the StepFun API for a limited time
StepFun is a Chinese AI lab with an API platform for its Step language and StepAudio speech models. Its StepAudio 3 series (speech recognition, text-to-speech, realtime voice, voice and music generation) is free on the API until 23:59 on November 8, 2026 (UTC+8), and billed at list prices from November 9.
Verified Nov 21 post on X