Scribe v2 for a 6-hour workshop with both add-ons vs Sume STT

Six hours of workshop audio costs $1.32 on ElevenLabs Scribe v2 at the standard $0.22 per hour, and $2.04 if you also turn on both listed add-ons ($0.050 per hour for keyterm prompting and $0.070 per hour for transcript editing). Sume STT 1.0 is $0.01 per audio minute, so the same six hours is 360 x $0.01 = $3.60. On price alone Scribe v2 is lower; Sume's case is the rest of the pipeline.

Vendor numbers come from the ElevenLabs API pricing page (read 2026-10-09). Sume numbers come from the Sume API catalog (read 2026-10-09).

The arithmetic

Standard transcription: 6 h x $0.22 = $1.32. Keyterm prompting adds 6 x $0.050 = $0.30, giving $1.62. Transcript editing adds 6 x $0.070 = $0.42, giving $1.74 with only that add-on. Both add-ons: $0.22 + $0.050 + $0.070 = $0.34 per hour, so 6 x $0.34 = $2.04.

Sume: 6 hours is 360 minutes, and 360 x $0.01 = $3.60. The extra is $3.60 - $2.04 = $1.56 against the fully loaded Scribe figure, or $2.28 against the base.

What Sume STT returns and where it stops

Sume's catalog describes STT 1.0 as priced per audio minute, with word timings always returned and optional sentence segmentation. There is no keyterm list and no transcript-editing add-on in the request schema: the fields are audio_url, language_code, segmentation, duration_seconds and metadata. If domain vocabulary is the main reason you would pay for add-ons, Scribe's add-ons answer a need Sume's STT does not.

A job also has a ceiling: the duration_seconds hint is capped at 600 seconds and, if you omit it, Sume reserves one minute. For a six-hour recording that means splitting the audio first. Audio detach can cut ranges out of a video at $0.01 per job.

When the Sume price is worth it

The $1.56 gap buys nothing if you only need a text file. It starts to be worth it when the transcript feeds another Sume step. The same account can detach audio from the video, transcribe it, and pass the words to the video captions job as words, which skips a second transcription. Decide on the pipeline, not the per-hour rate.

A decision rule

If the workshop is a one-off and the text file is the product, take the lower rate and spend the difference on a proofreader. If the recordings are video and you will also cut clips, burn captions and keep the audio for edits, price the whole chain instead of one line.

For a single six-hour recording the gap is a few dollars either way. The difference only becomes a budget item at volume: 100 hours a month is $22.00 on Scribe v2 standard against $60.00 on Sume STT, a $38.00 difference that you should weigh against the tools you would otherwise buy for the other steps.