Cost of 100 voiceovers by script length with Sume TTS

One hundred voiceovers of 300 characters each cost $1.425 on Sume text-to-speech, because the price is $0.0475 per 1,000 characters and a hundred 300-character scripts are 30,000 characters. Script length is the only input that moves the bill, so you can price a batch before you write a line.

The rate is Cartesia's list of $0.000038 per credit, at one credit per character, times the 1.25 house margin. The figures were read from the pricing package and API on 2026-10-03, and the length column assumes 15 characters per second of speech, which is a planning assumption you should replace with a measurement of your own voice and pacing.

The per-script price table

Spaces and punctuation count as characters. A request may carry up to 20,000 characters, so the largest single voiceover in the table costs $0.95.

Notice how little a short script costs. A 150-character voiceover is $0.0071, under a cent, and a hundred of them cost $0.7125.

A realistic mixed batch

Real batches are not uniform. Take sixty short product lines, thirty medium scripts and ten long explainers. The characters add up as below and the whole batch is one multiplication.

That is $2.28 for 100 voiceovers. Even if you generated every script twice to pick the better read, the batch would be $4.56.

Where the real money goes after the voice

Voice is almost never the expensive part. Captions are $0.20 per video of up to 60 seconds, so captioning the same 100 voiceover videos costs $20.00, which is more than 100 voiceovers at 600 characters ($2.85). Rendering each into a one-minute Timeline adds $10.00 for 100 minutes.

So when you budget a voiceover campaign, price the voices at cents and the finishing at dollars. The 100-video finishing stack is $20.00 captions plus $10.00 Timeline, $30.00 before any picture or clip.

• Voiceover: $0.0475 per 1,000 characters.
• Captions: $0.20 per job, up to 60 s.
• Timeline render: $0.10 per ceil(output minute).
• Music: $0.125 per generation.

How to check the batch before you run it

Count characters in code before you submit: sum the string lengths, multiply by 0.0000475, and compare with the wallet. That gives a cap you can refuse to cross. A hundred scripts at the 20,000 character limit would be $95.00, so the worst case for a batch is known up front, and so is the typical case.

A common mistake is pricing by the final audio length. Sume does not do that; two scripts that read at the same speed but differ in punctuation can cost slightly different amounts, because spaces and punctuation count as characters.

What Sume does and does not do

Sume TTS bills per character of the transcript, whatever the voice sounds like in the end. Check GET /v1/catalog for the live rate before quoting a client; the docs say the same for every public rate.

Rerun the arithmetic

The same line items as a short script, so you can change a quantity and see the total. It is plain arithmetic on the rates in this post; it does not call the Sume API.

# Sume billable USD rates, read 2026-10-03. Edit the quantities.
lines = [
    ('Voiceover characters, 100 x 300', 30000, 0.0000475),
]
total = 0.0
for label, qty, unit in lines:
    cost = qty * unit
    total += cost
    print(f"{label:<46}{qty:>8g} x {unit:<10g} = ${cost:,.2f}")
print(f"{'Total':<46}{'':>8}   {'':<10}   ${total:,.2f}")