Seed Audiov1

Seed Audio v1

Voice, music and sound effects from a single prompt.

Cost750

750 credits per run · 1 audio file · About 30 seconds

Explore by category

Popular

Newest

Seed Audio v1 audio generator

Type
Audio model
Made by
ByteDance Seed
Input
Optional text
Output
Audio
Price
750 credits per run
Plan
All plans, including Free

Seed Audio v1 is ByteDance Seed's audio model. Describe what you want to hear in a single text prompt, whether that is a spoken line, a piece of music or a sound effect, and it returns an MP3.

What you can do with it

  • Voice. Write the words to be spoken and describe how they should sound.
  • Music. Describe a mood, genre, instruments and tempo.
  • Sound effects. Describe a sound or an ambience: a door creaking, a busy train station, rain on a car roof.
  • Steer the voice. Attach up to three reference audio clips to guide how the voice sounds.

How to use it

  1. Write your prompt in the composer.
  2. Attach up to three reference audio clips if you want to steer the voice.
  3. Check the price in the composer, then generate. The result is an MP3.

Writing prompts for Seed Audio v1

There are no settings to adjust, so the prompt carries everything.

  • Say what kind of audio it is. Start with "A voice saying…", "Music:", or "Sound effect:" so the request is clear.
  • For voice, give the exact words in quotes, then describe the delivery: calm, excited, whispered, slow, warm.
  • For music, name the parts. Genre, instruments, tempo, mood and where it might be used, such as an intro or a background bed.
  • For effects, describe the space. A footstep in a cathedral sounds different from one on a carpet. Mention distance and surroundings.
  • Keep one idea per run. A single clear request is easier to judge and adjust than several mixed together.

Example prompts:

  • A warm, friendly voice saying: "Welcome back. Let's pick up where we left off." Unhurried, smiling.
  • Music: a gentle lo-fi beat with soft piano chords and vinyl crackle, relaxed and slow, for a study video.
  • Sound effect: a heavy wooden door creaking open in a large stone hall, with a faint echo.

Price and availability

Each run costs 750 credits, whatever you ask for. The composer shows the exact price before you generate. Seed Audio v1 is available on all plans, including Free.

FAQ

What is Seed Audio v1?

An audio model from ByteDance Seed. It makes voice, music and sound effects from a single text prompt.

What file do I get?

An MP3.

Can I guide the voice?

Yes. You can attach up to three reference audio clips to steer the voice.

Are there settings to adjust?

No. Everything is described in the prompt, so there are no extra options to set.

How much does it cost?

750 credits per run. The composer shows the exact price before you generate.

Do I need a paid plan?

No. Seed Audio v1 is available on all plans, including Free. Each run uses credits.