Transcribe your audio backlog
for $0.10 an hour

Send a list of audio URLs. Get back transcripts with sentence and word timestamps, as SRT, VTT, plain text or JSON. Whisper large-v3-turbo on a network of independent GPUs, built for backlogs rather than live captions.

How it works

Price per audio hour

List prices for batch or pre-recorded audio, from each vendor's pricing page as of October 2026. We are not the lowest per hour: Groq lists the same model for less, one file per request. We are built for backlogs: one call for many URLs, long files split for you, webhooks, and 100 free hours. Full comparison.

Get an API key

Credit

The API

curl https://api.myaitoken.io/v1/batches \
  -H "Authorization: Bearer $MYAI_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "job_type": "transcription",
    "model": "whisper-large-v3-turbo",
    "items": [
      {"input_url": "https://example.com/episode-101.mp3",
       "duration_s": 3540, "meta": {"language": "en"}}
    ]
  }'

# status (poll every 30 s, or pass "webhook_url" when you submit)
curl https://api.myaitoken.io/v1/batches/$BATCH_ID -H "Authorization: Bearer $MYAI_KEY"

# captions for item 0
curl "https://api.myaitoken.io/v1/batches/$BATCH_ID/results?format=srt&idx=0" \
  -H "Authorization: Bearer $MYAI_KEY" -o episode-101.srt

# JSON with sentence timestamps (add "word_timestamps": true to an item's meta for words)
curl "https://api.myaitoken.io/v1/batches/$BATCH_ID/results" -H "Authorization: Bearer $MYAI_KEY"

duration_s is required: it sets how long files are split and what is held from your credit. Results include timestamps: segment when sentence timing is available.