PicX Studio

AUDIOByteDance · from 15 credits / generation

ByteDance's speech model

Seed Audio 1.0 — AI Speech and Dialogue

20 preset voices. Speed, pitch, volume.

Seed Audio 1.0 turns text into natural speech and dialogue. Pick one of 20 preset voices and set the speed, pitch and volume.

Try Seed Audio 1.0See examples
  • Text

Natural speech, your way

20 preset voices

Choose from 20 voice presets. All of them speak Chinese, 16 also speak English, and a few also cover Spanish, Japanese, Indonesian or Portuguese, so one script can be voiced in more than one language.

Tune the delivery

Set the speed and volume from half to double, and shift the pitch up or down by up to an octave, so the same preset fits a calm explainer or a quick ad.

Choose the file

Download MP3 by default, or WAV, PCM or Ogg Opus at a sample rate from 8 to 48 kHz, ready for your editor or app.

Made for spoken audio

  • Voice-overs. Narration for product videos, explainers and social clips, at the speed and pitch you set.

  • Assistants and app prompts. Short, friendly lines like briefings, reminders and onboarding messages.

  • Bilingual content. Voice the same script in English and Chinese with one preset, or in Spanish, Japanese, Indonesian or Portuguese where the preset supports it.

Seed Audio 1.0

Why pick Seed Audio

Try Seed Audio 1.0
  • 15Credits per generationOne flat price, whatever the voice or settings.
  • 20Voice presetsChinese on all of them, English on 16, more languages on a few.
  • ±12Semitones of pitchShift the voice up or down by up to an octave.
  • 0.5–2×Speed and volumeSlow it down for clarity or speed it up for short ads.
  • 48 kHzTop sample rateMP3 by default, or WAV, PCM and Ogg Opus.

What Seed Audio 1.0 costs

One flat price per generation. The same price the generator shows before you run it.

  • Per generation
    15credits / generation
    On Starter≈ 240 generations / mo

Every model on every planCredits never expireSee plans

How to use Seed Audio 1.0 in 3 steps

  1. 1

    Pick a voice

    Choose one of the 20 preset voices.

  2. 2

    Write the script

    Type exactly what should be said.

  3. 3

    Tune and generate

    Set speed, pitch and volume if you need to, then generate. It costs 15 credits.

Seed Audio 1.0 prompts to copy

Real prompts and what they made. Copy one, or open it in the generator with the model already picked.

  • Daily briefing

    Here's a quick look at your day: you've got three meetings, one lunch, and just enough time in between to grab a coffee and relax.

    Try it

  • Product voice-over

    Meet the bottle that keeps your coffee hot for twelve hours and your water cold for twenty-four. One lid, one hand, zero spills.

    Try it

  • Two-line dialogue

    Did you lock the door? — Twice. And then I went back to check again.

    Try it

Seed Audio 1.0 questions

What is Seed Audio 1.0?

Seed Audio 1.0 is ByteDance's speech model. On PicX Studio it turns text into natural speech and dialogue in one of 20 preset voices.

How much does Seed Audio 1.0 cost?

15 credits per generation.

Which languages does it speak?

All 20 presets speak Chinese and 16 of them also speak English. A few also speak Spanish, Japanese, Indonesian or Portuguese.

Can I upload a voice sample?

No. On PicX Studio, Seed Audio speaks in its preset voices only. We don't take reference recordings, so it can't be used to imitate a real person's voice.

What can I adjust?

Speed and volume from 0.5 to 2 times, pitch from 12 semitones down to 12 up, the output format (MP3, WAV, PCM or Ogg Opus) and the sample rate, from 8 to 48 kHz.

Can I add a Seed Audio voice to a video?

Yes. Generate the speech, then sync it to a talking character with VEED Lip Sync, or use it as a voice-over in your editor.

Give your script a voice.

Generate natural speech and dialogue with Seed Audio 1.0 on PicX Studio, 15 credits per generation.

FREEClaim Free Image Generation