On Sept 15, foundation model company StepFun launched its next-gen StepAudio 3 series, unveiling five models at once: StepAudio 3 Realtime, StepAudio 3 ASR, StepAudio 3 TTS, StepAudio 3 Gen, and StepAudio 3 Music.

They cover real-time voice interaction, speech recognition, human-level voice generation, and music creation. Several models ranked No.1 globally on Artificial Analysis, an authoritative third-party AI model evaluation platform.