Voiceover cost for a 60-second video: about $0.043 on Sume TTS

A 60-second voiceover costs about $0.043 on Sume TTS. That assumes a script of around 900 characters, a budgeting figure I use here (roughly 15 characters per second including spaces), not a measured promise. Sume bills characters, so count your actual script rather than guessing from the runtime.

Script length to bill

Spaces and punctuation count, and the maximum is 20,000 characters per request. Sume TTS is $0.0475 per 1,000 characters, so the figures below are 0.0475 times the character count in thousands.

Count before you submit

Measure the exact string you will send. It takes one line:

import os, requests

script = open("script.txt", encoding="utf-8").read().strip()
print(len(script), "characters")
r = requests.post("https://api.sume.com/v1/tts-1.0/generate",
    headers={"Authorization": "Bearer " + os.environ["SUME_API_KEY"],
             "Idempotency-Key": "vo-60s-001"},
    json={"transcript": script, "voice": {"id": os.environ["VOICE_ID"]}})
print(r.status_code, r.json())

The other costs in a 60-second video

The voice is the smallest line. A one-minute timeline render is $0.10, a captions job is $0.20, and a music bed is $0.125. Voice plus those three is about $0.468 before the footage.

The request needs a voice.id in the library-id shape; made-up names such as a voice label are refused at the door. The Microsoft figures, $22 per 1M characters for MAI-Voice-2.1 and $15 for MAI-Voice-2.1-Flash (read 2026-10-08), come from Microsoft AI; they are list prices for different products.