Amazon Transcribe $0.006 per minute vs Sume STT reservation
Amazon Transcribe bills batch audio at $0.006 a minute in one-second steps. Sume STT is $0.01 a minute and reserves one minute unless you pass duration_seconds.

Amazon Transcribe's pricing page lists batch transcription at $0.006 per minute in the US East (N. Virginia) region, billed in one-second increments with no minimum. Sume STT 1.0 is $0.01 per audio minute. Amazon is cheaper at list price, and its volume tiers go lower.
The practical difference for short clips is how the amount is reserved: Sume reserves one minute when you omit duration_seconds, so a 10-second clip should send a hint.
Published figures
Amazon's figures are for the first pricing tier in one region as printed on its page; other regions and features differ.
| Item | Amazon Transcribe | Sume STT 1.0 |
|---|---|---|
| Batch rate | $0.006 per minute (US East N. Virginia) | $0.01 per audio minute |
| Billing increment | One second, no minimum | Reserved from duration_seconds; default 1 minute |
| Volume tiers | First 250,000 minutes, then 38% off for the next 750,000, then 58% off beyond 1M | One public rate |
| Free tier | 60 minutes per month for 12 months | Not claimed |
Make Sume reserve what you need
Pass duration_seconds between 1 and 600 so the reservation matches the clip. The cap is per request, so split recordings longer than ten minutes before you submit.
import os, requests
r = requests.post(
"https://api.sume.com/v1/stt-1.0/transcribe",
headers={"Authorization": f"Bearer {os.environ['SUME_API_KEY']}",
"Idempotency-Key": "stt-demo-001"},
json={"audio_url": "https://example.com/call.mp3",
"language_code": "en",
"duration_seconds": 300},
)
print(r.status_code, r.json())Which to pick
For a very large archive of plain transcripts inside AWS, Amazon's per-second billing and tiers are hard to beat on price. For transcripts that feed other Sume steps (captions, dubbing, timeline audio), the single job model avoids moving files between clouds.
Neither number is a quality comparison; run a sample of your own audio through both.
Worked example
A one-hour archive and a 20-second clip show the difference between the two billing approaches.
- 100 hours at $0.006 per minute: Amazon $36.00 versus Sume $60.00.
- A 20-second clip: Amazon bills 20 seconds (about $0.002). Sume reserves one minute ($0.01) if
duration_secondsis omitted, so send the hint. - Amazon's tiers (38% off the next 750,000 minutes, 58% beyond 1M) matter only at very large volume.
Checklist before you commit
Amazon's printed rate is for one region and its first tier. If you run outside US East (N. Virginia), read your own region's page.
- Check for an AWS-only data requirement before comparing prices.
- Check that your clips fit inside 600 seconds, or split them first.
- Compare accuracy on your own audio, not just price.
Sources
Related posts
More in Comparisons
- Argil audioUrl is limited to 40 seconds; what Sume takes instead
Argil's moment audioUrl is capped at 40 seconds. Sume's avatar talking-video takes script text or silence beats, not an audio file. What that means.
- Argil 500-character moment limit vs Sume's 4-60 second script window
Argil caps each moment transcript at 500 characters; Sume sizes a script by estimated seconds instead. How to split a long avatar script on either one.
- Argil subtitles Top/Middle/Bottom and sizes vs Sume caption design
Argil subtitles take a styleId, a position and a size. Sume burns captions with a named style plus design overrides. Compare the knobs before you pick.
- Argil video status IDLE to DONE vs Sume queued to completed
Argil videos move through IDLE, GENERATING_AUDIO, GENERATING_VIDEO, DONE or FAILED; Sume jobs go queued, processing, completed. How to map polling code.
Written by Sume