ai.neiroset.com

Seed Audio 1.0: AI voice for scenes and dialogue

Seed Audio 1.0 voices a scene, not just text: describe where and how the lines are spoken, and the model reads them with the right delivery.

  • ByteDance
  • Seed Audio 1.0
  • scenes
  • dialogue

How it works

[whispers]<#1.5#>

Step 1

Paste your text

An article, a video script or a dialogue. Pauses and audio tags are added with the buttons under the box.

BriannarratorJessicabrightHankstoryteller

Step 2

Pick a voice

Voice, delivery, emotion and speed: all in the tiles under the box.

Step 3

Download the MP3

The audio lands in your history: listen here and download the file.

What Seed Audio 1.0 can do

Seed Audio 1.0 is a speech model from ByteDance. Unlike regular text to speech, it understands a scene description: where the conversation takes place, who is speaking and in what mood. Cues like “tired”, “delighted” or “busy cafe” are not read aloud; the model turns them into delivery. On ai.neiroset, Seed Audio works next to other AI models in one account.

  • A scene instead of bare text. A description of the place and the mood sets how the lines sound.
  • Stage directions stay silent. “Tired”, “in a whisper”, “delighted” change the delivery instead of ending up in the audio.
  • English and other languages. For text that is not in English, the model switches to its multilingual mode automatically.
  • Your own voice. Record or upload 10–15 seconds of speech, and Seed Audio reads in your voice for 12 months.
  • Speed and pitch. Speed from 0.5x to 2x, pitch and volume in Settings.

Seed Audio 1.0 specs

Developer
ByteDance
Model
Seed Audio 1.0
Voice
Picked by the model from the scene, or your own
Text per run
Up to 1,500 characters
Speed
0.5x to 2x
File
MP3 or WAV
A short scene
Ready in about 20–25 seconds

How to voice a scene in Seed Audio

  1. Step1
    Describe the scene
    In one sentence: where the conversation happens and what the mood is.
  2. Step2
    Write the lines
    Before each one, say who speaks and how: “A man, tired”, “A girl, in a whisper”. Put the words themselves in quotes.
  3. Step3
    Click “Generate voice”
    The price shows on the button before you run it. Download the finished scene as MP3 or WAV.

Example scenes to voice

Seed Audio free

The 2 free voiceovers you get after signing in work for Seed Audio too, up to 300 characters each.

  • If a voiceover fails, the attempt comes back automatically.
  • After that, voiceovers are paid with tokens per character, and the price shows on the button before you run it.

Seed Audio FAQ

No. ai.neiroset is an AI aggregator and is not affiliated with ByteDance. Seed Audio runs here next to other voice models.

There is no list of ready voices: the model picks the voice and manner from the scene. You can also create your own voice for Seed Audio. For a fixed voice from a list, use MiniMax or ElevenLabs. MiniMax Speech 2.8 HD

No. Words like “tired” or “delighted” before a line set the delivery, and only the lines themselves are spoken.