Seed AudioSeed Audio

AI voice models

Every generation on this site uses Seed Audio 1.0 — an audio model with text-to-speech and voice cloning, 20 preset voices, multilingual output, and speed, pitch and volume controls.

Generate voiceView pricing

Models

Generate voice from text or reference audio for your use case.

Model specs

Key Seed Audio 1.0 specs.

ModelInputOutputFormatsVoicesCloningCreditsBest for
Seed Audio 1.0Text, reference audio (up to 3)Audio fileMP3 / WAV / PCM / OGG (up to 48 kHz)20 presetsYes (30s each)Usage-based (shown upfront)Narration, TTS, voice cloning

Pick by use case

When to use Seed Audio 1.0.

Narration & read-aloud

Seed Audio 1.0

Natural, expressive text-to-speech for video and explainer narration.

Generate in your own voice (cloning)

Seed Audio 1.0

Reproduce a voice from reference audio (up to 3 clips, 30s each).

Podcasts & audiobooks

Seed Audio 1.0

Reads long scripts with steady tone that stays consistent across chapters.

News & announcements

Seed Audio 1.0

Choose voices and speed suited to a clear, steady read.

Multilingual content

Seed Audio 1.0

Supports English, Chinese, Japanese, Spanish, Indonesian and Portuguese, including mixed-language text.

Character voices

Seed Audio 1.0

Pick from 20 presets to match the tone and use case.

Model FAQ

Common questions about Seed Audio 1.0.