40 speaker intro videos: avatar creation plus 15 seconds each on Sume

Forty conference speaker intros of 15 seconds each cost $148.40 on Sume Avatar 1.0 Standard, $185.20 on Plus and $368.00 on Max, if each speaker needs a new avatar. The numbers are $0.95 to create an avatar from a reference image and $2.76, $3.68 or $8.25 for a 15-second video at $0.184, $0.245 or $0.55 per second without a product (read 2026-10-06 on the API rate card).

The creation fee is a one-time cost per speaker. If the same speakers return next year, the second year is the video line alone: $110.40 on Standard for 40 more intros. If you already hold avatars for a roster, the first year looks like the second.

Standard, Plus or Max for a speaker card

A short intro on a conference page is watched small, and a faster, cheaper tier is a defensible choice for the many. Max costs three times as much as Standard, so a mixed plan makes sense: Standard for the 30 supporting speakers, Max for the 4 keynotes, and Plus for the other 6. That plan is 30 × $3.71 + 6 × $4.63 + 4 × $9.20 = $175.88, against $148.40 for all Standard.

The video line scales with seconds, so cut the script before you upgrade the tier. A 12-second intro on Standard is $2.21 and costs $0.55 less than a 15-second one.

• Get written permission from each speaker before you create an avatar from their photo.
• Name each avatar handle after the speaker's session slug, so the roster is searchable in your own records.
• Keep the intro script at 30 to 40 words, which is about what 15 seconds of speech holds.
• Read GET /v1/balance before you start: a Standard roster needs $148.40 available, and a smaller balance returns 402 insufficient_credits.

Price the roster by tier

The function takes a roster as tier counts and seconds, and returns the total including creation. Change the counts to cost your own mix.

RATE = {"standard": 0.184, "plus": 0.245, "max": 0.55}
CREATE = 0.95


def intro_cents(tier: str, seconds: int) -> int:
    return -(-round(RATE[tier] * seconds * 1000) // 10)


def roster(counts: dict, seconds: int = 15) -> float:
    cents = sum(n * (95 + intro_cents(t, seconds)) for t, n in counts.items())
    return cents / 100


print(roster({"standard": 40}))
print(roster({"standard": 30, "plus": 6, "max": 4}))

Captions and what the totals leave out

The totals above leave out captions. If you burn captions onto the finished intros as a separate step, the standalone caption job is $0.20 per video for clips up to 60 seconds, so 40 intros add $8.00. Inline captions on an avatar video are a separate add-on in the avatar-video estimate and do not create a separate caption job, so read the live price in the catalog before you rely on either number.

Also leave room for a second take on the script. A speaker who does not like their intro is one more 15-second video, $2.76 on Standard, and the avatar is already made.

If the speaker list changes

Speaker lists move until the week of the event. A speaker who drops out costs you nothing if you have not created the avatar yet, and $0.95 if you have. A speaker who is added late costs $3.71 on Standard. So create avatars last, after the programme is confirmed, and generate the intros in two waves: the confirmed speakers first, the late additions after.

A 20 percent reserve covers the usual churn. On a $148.40 Standard roster that is $29.68, so a $178.08 balance covers the roster and the changes. If the balance falls short, the submit fails with 402 insufficient_credits before any work starts, and nothing is charged.

After the first year

Avatars are reusable handles, so the creation fee does not repeat for a returning speaker. Budget the first event at the full number and the next at the video line. Intros for a 40-speaker event are a small line next to the cost of the event itself, but the roster should still be approved as a number before the first job runs. The tier prices are on the API pricing page and the video parameters are in the avatar video docs.