About Seed Audio

Seed Audio is a service for generating natural AI speech from text or reference audio. It is powered by Seed Audio 1.0, ByteDance's audio generation model, and supports text-to-speech (TTS) and voice cloning.

It offers 20 preset voices, speed and pitch control, and export in several formats, with the credit cost always shown before you generate. The goal is to keep model choice, settings, and generation in one clear workspace.

For product, billing, or privacy questions, contact support@seedaudiojp.com.