How much does closed captioning cost? Per clip and per minute

What closed captioning costs depends on who or what writes the captions and how the price is counted: per minute of video, or per clip. Automated captions on Sume cost $0.20 per clip of up to 60 seconds, so about $0.20 per minute of video, but they are burned into the picture (open captions). A separate closed-caption file comes from a transcript billed at $0.01 per audio minute.

The Sume prices below are read from the code that sets them and match the Video captions, Video trim, Audio detach, Timeline 1.0, and Video inspect docs, read on 2026-09-29. Anything described as current behavior is read from Sume's code. This post prices Sume's automated captions only, not human or live captioning.

Are these closed captions or open captions?

Open. Sume's caption job burns the words into the picture and returns a new captioned video, so the captions always show. Closed captions are a separate track the viewer can turn off; hard vs soft subtitles explains the difference and how to make the separate file.

The caption job does not hand back a caption file: the docs say raw transcripts are not part of its public contract. If you need a closed-caption file, the cost is a transcription instead, billed per audio minute (see the table below), and you write the file from the timings.

What does one caption job cost?

Each accepted standalone caption job reserves and captures $0.20, for videos up to 60 seconds, under the current fixed estimate. The docs say to confirm live pricing in GET /v1/catalog. Every job is billed plus a 5.5% agent fee by default.

• The input is one public HTTPS video URL. Localhost, private-network, signed or private URLs are rejected.
• In current code the caption job refuses a source longer than 60 seconds, and a source with no audio stream.
• The price is per job, not per word, so a 20-second clip and a 60-second clip cost the same.

How much does captioning a 10-minute video cost?

A video over 60 seconds is captioned in pieces: cut it into chunks of 60 seconds or less, caption each chunk, then rejoin them over the original sound. The method is in Add captions to a long video by API. For a 10-minute video cut into 10 chunks of 60 seconds, that is $2.00 in captions and at most $3.31 for every step before the agent fee. Each extra chunk, for example from cutting at sentence ends, adds one trim job and one caption job.

The cut, detach and render jobs reserve their listed rate and capture their own compute, never above the reservation, so those rows are ceilings. They also read only files already in your Sume workspace, not any public URL.

What does a caption file cost instead?

Transcription is billed per audio minute: $0.01 per audio minute for STT 1.0, which takes up to 10 minutes per request. On a clip already in your Sume workspace, video inspect with transcribe: true bills at the same rate and can return sentence segments[] with start and end, which map onto numbered caption blocks. The video itself is not changed, so there is no render charge.

What is not in the price?

• Human review, accuracy checks, and translation. The caption job writes what speech-to-text hears, or the words and times you send as cues.
• Live captioning of a meeting or stream. That is a different product and is not priced here.
• Retries you trigger yourself: a new caption job is a new charge. Restyling a finished caption with source_caption_id skips a second transcription, but the docs say billing is unchanged, because a restyle is still a render.