Bedtime story video with a lullaby bed: Sume TTS, Lyria, timeline cost

Cost of a narrated bedtime story with a lullaby underneath: TTS for 3,000 characters, one Lyria 3.5 track, and a timeline render with soundtrack ducking.

5 min readSume
All posts

Short answer

A narrated bedtime story with a lullaby bed is three paid steps on Sume: narration, music and the render that mixes them. For an invented 3,000-character story that runs about 4 minutes 20 seconds, the narration is 15 cents, one Lyria 3.5 generation is 12.5 cents, and a 5-minute render is 50 cents, so about 77.5 cents for the first version.

Everything here is a worked example. The story length and duration are made up; the prices come from the Sume docs and the TTS rate.

Cost of one pass of the example story (Sume public rates, read 2026-10-07)
StepBasisCost
Narration (TTS)3,000 characters x $47.50 per 1M = $0.1425, rounded up per job$0.15
Lullaby bed (Lyria 3.5)one generation, fixed Music price$0.125
Render (Timeline 1.0)ceil(260 s / 60) = 5 minutes x $0.10$0.50
Total$0.775

Making the bed

Music cannot be given a duration: Sume rejects duration and duration_seconds, so put the length in the prompt and make the track at least as long as the story, or set loop on the soundtrack. The prompt should ask for a gentle instrumental with no vocals, since the voice already carries the words.

The render

The render takes the narration as the audio spine, with audio.duration_seconds set to the narration length, and the music as soundtrack. The soundtrack fields in the docs are url, gain_db, loop, fade_out_seconds up to 10 and duck_db from 0 to 20. A duck of a few decibels keeps the voice clear, and a 10-second fade lets the last page end softly. The video side is a list of picture slots; the first must start at 0 and each must last at least 0.2 seconds.

{
  "audio": { "url": "https://media.sume.com/artifacts/artf_demo/story.wav", "duration_seconds": 260 },
  "soundtrack": { "url": "https://media.sume.com/artifacts/artf_demo/lullaby.wav", "gain_db": -12, "loop": true, "fade_out_seconds": 10, "duck_db": 8 },
  "video": [
    { "source_url": "https://media.sume.com/artifacts/artf_demo/page1.png", "start": 0, "duration": 130 },
    { "source_url": "https://media.sume.com/artifacts/artf_demo/page2.png", "start": 130, "duration": 130 }
  ]
}

Preview the price

Use POST /v1/timeline-1.0/plan first. It is unbilled and returns the duration and billable minutes, so a 260-second story shows 5 before you spend. Note that 260 seconds is just over 4 minutes, and a story of 240 seconds would bill 4 minutes, so trimming 20 seconds saves 10 cents.

Changing one part

Voice choice and picture style are up to you. Re-rendering with a different bed costs one more render, but not another narration.

Reuse across a series

A series reuses the same lullaby for every episode. One Lyria generation at 12.5 cents is then paid once, and each new episode adds only its narration and render. With a 260-second story the per-episode cost drops from 77.5 cents to 65 cents, since the music is already on media.sume.com.

Keep the soundtrack file in your workspace and point each render at it. Set loop to true when the bed is shorter than the story, and a 10-second fade_out_seconds ends it gently.

Before you publish anything for children, listen to the whole mix at bedtime volume. A duck_db that sounds fine at a desk can feel too loud once the room is quiet.

  • Pay for the lullaby once, reuse the file.
  • Set loop for shorter beds.
  • Listen at the volume the story will be played at.

Sources

Related posts

More in Use cases

All Use cases posts

Written by Sume