36 slides at 700 characters: about $1.20 of Sume TTS
Sume TTS bills $0.0475 per 1,000 characters. 36 slides of 700 characters is 25,200 characters, about $1.20, or about 3 cents a slide.

Thirty-six slides at 700 characters each is 25,200 characters. At Sume's $0.0475 per 1,000 characters that is 25.2 x $0.0475 = $1.197, or about $0.0332 per slide. Spaces and punctuation count toward the characters, so measure the real script, not the word count.
The rate and its limits
The API pricing page lists Text to speech as priced per transcript character after Sume margin. The catalog description in the repo gives a maximum of 20,000 characters per job, an estimate of 1,000 characters when the length is unknown, and a minimum charge of 1 cent. A single 25,200-character job would be over the limit, which is one more reason to make one job per slide.
Per-slide jobs
The MCP tools page describes the same shape for agents: one tts_create per sentence, one image per scene. One job per slide also means a bad take costs one slide, not the deck.
| Slide length | Characters per deck | Per-slide cost | Deck cost |
|---|---|---|---|
| 300 characters | 10,800 | $0.0142 | $0.513 |
| 500 characters | 18,000 | $0.0238 | $0.855 |
| 700 characters | 25,200 | $0.0332 | $1.197 |
| 1000 characters | 36,000 | $0.0475 | $1.71 |
Retakes and the one-cent floor
Plan for retakes. If one slide in four is re-recorded once, 9 extra 700-character jobs add 9 x $0.03325 = $0.29925, which brings the deck to about $1.50. A very short slide of 100 characters would compute to $0.00475, but the minimum is 1 cent, so a deck of many tiny captions costs more per character than a deck of long paragraphs.
Compare the plan against the number of jobs rather than the minutes of audio. Characters set the bill; the spoken length only matters for your timeline.
- Count characters in the final text, including spaces and punctuation.
- Keep each job under 20,000 characters.
- Budget for retakes separately from the first pass.
Estimating from words, not characters
If you only have word counts, English prose averages about 6 characters per word including the space, so 700 characters is roughly 115 words. That is a rule of thumb, not a Sume figure, and punctuation-heavy scripts run higher. Count the final strings in code before you submit and use that number for the budget.
The estimate you submit with matters too. The catalog description says that the estimate is 1,000 characters when the length is unknown, so sending the real transcript keeps the reserve close to the real price. A 700-character slide reserves about 3 cents instead of 5.
- A 36-slide deck at 1,000 characters a slide is 36,000 characters and $1.71.
- A 20-slide deck at 700 characters is 14,000 characters and $0.665.
- Doubling the deck length doubles the bill; the per-job floor is the only non-linear part.
Sources
Related posts
More in Pricing
- 40-second AI presenter video: $7.36, $9.80 or $22.00 on Sume
A 40-second talking avatar video bills 40 seconds at the tier rate: $7.36 Standard, $9.80 Plus, $22.00 Max, before any product image surcharge.
- 402 on an 8-second 720p Gemini Omni clip: the balance must cover $1.00
Sume reserves the billable amount at submit and returns 402 insufficient_credits if the balance is short. One 8 s 720p Omni clip needs $1.00; 25 need $25.00.
- 45,000-character chapter on Sume TTS: split evenly costs 2 cents more
A 45,000-character chapter needs three TTS requests. Splitting 20,000 + 20,000 + 5,000 costs $2.14; an even 15,000 x 3 split costs $2.16.
- A 45,000-character chapter needs three Sume TTS jobs: about $2.14
Sume TTS caps a job at 20,000 characters. A 45,000 character chapter splits 20,000 + 20,000 + 5,000 and costs 45 x $0.0475, about $2.14.
Written by Sume