Arabic speech to text API: Sume STT with language_code ar

To transcribe Arabic speech with Sume, submit the audio to POST /v1/stt-1.0/transcribe with language_code set to ar as a hint, wait for the job to complete, and read text and words[] from the result. Sume prices STT at $0.01 per audio minute, so a 90 second Arabic clip reserves and bills about one and a half cents of audio time when you pass duration_seconds.

One honest caveat first. The Sume docs do not publish a list of supported languages for STT, so this page does not promise accuracy for Arabic. It shows how to send the hint, how to check what came back, and how to judge a sample before you commit to a batch.

The request

Use ar as the hint for Arabic. Spoken Arabic varies a lot by region, and the Sume docs give no dialect guidance, so treat the hint as a starting point and check the result against a sample from your own speakers. Try the clip with and without the hint.

The job is asynchronous. Submit returns 202 with a request_id, which is the job id. Poll GET /v1/jobs/{id}/status until it is completed, failed or canceled, then read GET /v1/jobs/{id}/result. Always send an Idempotency-Key so a retried submit does not create a second paid job.

curl -X POST https://api.sume.com/v1/stt-1.0/transcribe \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: stt-ar-001" \
  -d '{
    "audio_url": "https://media.sume.com/artifacts/artf_demo/clip.wav",
    "language_code": "ar",
    "duration_seconds": 90,
    "segmentation": "sentence"
  }'
# then poll GET /v1/jobs/$REQUEST_ID/status and read /result

Check the result before you trust it

Arabic text runs right to left, so check how the transcript renders where you store it. The text is a plain string and the words[] entries are in speech order with start and end seconds. Keep those timings as numbers, and only reverse or shape text in the display layer, not in the data you store.

Compare the language_code in the result with the one you sent, and read language_probability. If they disagree, the clip may be mixed-language, very short, or mostly music. Run two or three of your own recordings, including a noisy one, and read the text yourself or have a speaker of the language read it.

Captions from Arabic speech

Burned-in Arabic captions are not something the Sume captions docs cover: font coverage is documented for Latin and Hangul only, as explained in the stored note on Japanese, Chinese and Arabic captions. Use the transcript and word times for your own player's subtitle track, and verify with a short test before you plan a caption render.

Cost and limits

One STT job takes at most 600 seconds of audio. For a longer recording, split it into slices, send one job per slice, and add each slice's start time to its word times when you merge them. Splitting a video's audio first with audio detach costs $0.01 per job and gives you a 16 kHz mono WAV.

If you leave duration_seconds out, Sume reserves one minute, which is wrong for a long file. Pass the real length rounded up. Microsoft's current model page lists 60 languages for its transcription model (read 2026-10-07), which is a vendor claim about that model and not a statement about Sume's STT.

The result also has segments[] when you ask for segmentation, and each segment carries start and end seconds. For subtitles you do not need to cut the text yourself: pass the transcript or your edited script to the captions endpoint and let it keep the speech timings.