Sume STT takes 10 minutes per request: a 90-minute file is 9 jobs
Sume speech-to-text caps a request at 600 seconds of audio, so a 90-minute recording is nine requests at up to $0.10 each, $0.90 in total.
Sume speech-to-text accepts up to 10 minutes of audio per request, so a 90-minute recording needs nine requests. At $0.01 per audio minute each full request costs $0.10, so the whole file is $0.90.
| Recording | Requests of 10 min | Sume STT cost |
|---|---|---|
| 10 min | 1 | $0.10 |
| 45 min | 5 | $0.45 |
| 90 min | 9 | $0.90 |
| 180 min | 18 | $1.80 |
Split before you send
Cut the audio into chunks of 600 seconds or less, for example with the Timeline trim operation described in the Timeline docs, and send each chunk as its own request. The per-minute rate is on the rate card.
Stitching the transcript
Keep chunk order in your own code and join the text by index. Words that cross a boundary may split, so cut at silences where you can. The Video Inspect docs describe the STT job.
Price check
Total cost does not change with chunking: 90 minutes at $0.01 a minute is $0.90 either way. Microsoft's streaming model lists $0.54 an hour for the intro period, per its announcement, so the same file lists at $0.81.
Sources
Related posts
More in Developers
- Turn Sume STT word timings into an SRT file in 30 lines of Python
Sume speech-to-text returns words[] with start and end seconds. Group them into SRT cues of 42 characters or 5 seconds, and see what the caption endpoint takes.
- Count Sume job replays with idempotency_hit after a client retry
A Sume job submit returns data.idempotency_hit. Log it after a retried video submit to prove the second call replayed the first job, not a new one.
- Sume submit budgets: 120 to 1,200 writes a minute for an Omni batch
Sume rate limits submits per plan: Free 120, Pro 300, Startup 600 and Scale 1,200 a minute; reads get 40 times that. Why queue size, not the rate, paces Omni.
- Sume sync mode waits at most 30 seconds: short TTS vs long scripts
mode sync is a bounded wait, clamped to 0-30 seconds, not a promise the audio is ready. When to use it, and when to go async or webhook.
Written by Sume