Sign up and save up to 50% on a yearly plan.Sign up free →Black Friday: up to 50% off yearly plans.See the deal →

Speech from ByteDance Seed.

Seed Audio 1.0 AI voice generator

ByteDance Seed's audio creation model reads your script in 20 languages. Tune the speech rate, pitch and loudness, and get a WAV ready for your edit.

From $0.14 per 1,000 characters on MAX yearly

  • 20 languages
  • Up to 2,900 characters a read
  • WAV out

● Made with Seed Audio 1.0

Hear Seed Audio 1.0

Every sample is one read, with the script that made it and what it cost.

  • “Good evening, and welcome aboard. Tonight's flight to Lisbon will take about two hours and forty minutes. Sit back, relax, and enjoy the view.”

    Default voice
    142 characters · 5 credits
  • “New in the shop this week: hand-thrown mugs in four glazes. Each one is a little different, and that is exactly the point.”

    Default voice
    122 characters · 5 credits
  • “Here is today's forecast. Sunny in the morning, clouds by noon, and a chance of rain after five. Bring a jacket if you are heading out tonight.”

    Default voice · faster
    143 characters · 5 credits

● Why people pick it

What is Seed Audio 1.0 good at?

01

An audio creation model

ByteDance Seed built it to make speech, sound effects and ambience in one framework, not just a plain voice.

02

20 languages

English, Chinese, Japanese, Korean, Spanish, German, French, Portuguese, Indonesian, Thai, Vietnamese and more.

03

Rate, pitch, loudness

Three dials on the read itself: faster or slower, higher or lower, louder or softer.

04

Long, natural reads

Up to 2,900 characters a read on OpenClips, and up to two minutes of audio by Seed's own limit.

● Use cases

What people make with Seed Audio 1.0

Voiceovers

BytePlus names voiceovers, audiobooks and gaming as the scenarios Seed Audio is for.

Announcements

Calm, clear reads for travel, retail and events.

Localised video

One script read in the languages Seed covers, from Spanish to Thai.

● Side by side

How does Seed Audio 1.0 compare?

The other voice models on OpenClips, on the same credits.

SpecSeed Audio 1.0MiniMax Speech 2.8 HDElevenLabs v3ElevenLabs v4
Made bySeed Audio 1.0ByteDance SeedMiniMax Speech 2.8 HDMiniMaxElevenLabs v3ElevenLabsElevenLabs v4ElevenLabs
Price per 1,000 charactersSeed Audio 1.034 creditsMiniMax Speech 2.8 HD20.4 creditsElevenLabs v316.3 creditsElevenLabs v416.3 credits
A 1,000-character readSeed Audio 1.034 creditsMiniMax Speech 2.8 HD21 creditsElevenLabs v317 creditsElevenLabs v417 credits
Longest scriptSeed Audio 1.02,900 charactersMiniMax Speech 2.8 HD5,000 charactersElevenLabs v35,000 charactersElevenLabs v410,000 characters
VoicesSeed Audio 1.0By voice idMiniMax Speech 2.8 HDBy voice idElevenLabs v320ElevenLabs v420
OutputSeed Audio 1.0WAVMiniMax Speech 2.8 HDRASTERElevenLabs v3RASTERElevenLabs v4RASTER

Prices are the catalog's, quoted before every render.

● Specs

Seed Audio 1.0 specs

StatusLive in the app, the API and the MCP
Made byByteDance Seed
ModalityText to speech
InputsText script, one reference audio
Longest script2,900 characters
Languages20 languages
Loudness rate-50–100
Pitch rate-12–12
Sample rate8000 · 16000 · 22050 · 24000 · 32000 · 44100
Speech rate-50–100
Longest output (vendor)120 seconds a read
Model typeAudio creation: speech, sound effects and ambience (ByteDance Seed)
OutputWAV audio
Price34 credits per 1,000 characters · $0.14 per 1,000 characters on MAX yearly
A 1,000-character read34 credits

● How it works

How do you use Seed Audio 1.0 on OpenClips?

01

Describe it

Write the script, up to 2,900 characters. Every word is read aloud.

02

Pick the settings and generate

Choose the settings Seed Audio 1.0 offers, then generate. The result lands in your library.

● Tips

Tips for better results with Seed Audio 1.0

  • Change the speech rate

    Raise the speech rate for forecasts and recaps, lower it for stories.

  • Keep reads under two minutes

    Seed generates up to 120 seconds of audio a read. Split a long script into scenes.

  • WAV for editing

    The read comes back as WAV, lossless for mixing under music.

● Made for

What can you make with Seed Audio 1.0?

VoiceoversAudiobooksGame contentAnnouncementsPodcastsLocalised adsShort dramaNarration

● Questions

Seed Audio 1.0 FAQ

ByteDance Seed's audio creation model. Seed describes it as a model that generates speech, sound effects, ambience and other scene-level audio in one framework. OpenClips uses it to read scripts aloud.

BytePlus lists 20: English, Chinese, Japanese, Korean, Mexican and Castilian Spanish, Indonesian, German, Brazilian Portuguese, French, Thai, Vietnamese, Malay, Filipino, Italian, Russian, Dutch, Polish, Turkish and Swedish.

Seed Audio 1.0 opens with your first purchase: any plan or credit pack. A new account still gets ⚡30 on signup with no card, for the models the free plan runs.

Seed Audio 1.0 costs 34 credits per 1,000 characters, paid from your OpenClips credits. That is from $0.14 per 1,000 characters on MAX yearly. A 1,000-character script costs 34 credits. The charge is the characters you send, rounded up once.

Up to 2,900 characters a read. A longer script is refused before any credit is held, so split a long script into reads.

An WAV file, saved to your library, ready to download or to drop on a video.

Per 1,000 characters: Seed Audio 1.0 34 credits, MiniMax Speech 2.8 HD 21 credits, ElevenLabs v3 17 credits, ElevenLabs v4 17 credits. The table above sets their scripts, voices and files side by side.

Yes. Seed Audio 1.0 runs from the OpenClips app, the REST API and the MCP server, with the same settings and the same price.

● The receipts

Where do these Seed Audio 1.0 facts come from?

Every link goes to the vendor's own page.

Read your script with Seed Audio

20 languages, with rate, pitch and loudness control.

Generate with Seed Audio 1.0

What Seed Audio 1.0 does

Seed Audio 1.0 is ByteDance Seed's audio creation model. Seed built it to generate speech, sound effects and ambience in one framework. On OpenClips it reads a script of up to 2,900 characters and saves a WAV to your library.

Speech rate, pitch and loudness

Three settings shape the read: the speech rate, the pitch and the loudness. Leave them at their defaults for a neutral read, or move them for a faster forecast, a deeper narrator or a softer aside.

Seed Audio vs the other voices

Seed Audio costs more per character than the other voices on OpenClips. For many named voices, look at ElevenLabs v4 or Deepgram Aura-2.