Skip to content
/
Create transcription
Last updated

Create transcription

Submit an audio file for transcription by providing an HTTPS URL. The API pre-charges synapses based on audio duration, submits the job for processing, and returns an initial response with a transcription ID for polling.

POSThttps://api.aitronos.com/v1/audio/transcribe

Creates a new transcription job. The audio file must be accessible via a public HTTPS URL. The service validates the URL, calculates the synapse cost based on audio duration and any applicable multipliers, pre-charges the organization's balance, and submits the job for processing. The transcription starts in queued status and transitions through processing to completed or failed.

Billing uses consumption-based pricing with the FVT-1 model. Priority levels (normal, high, urgent) apply multipliers to the base cost.

Request Body

organization_id string required

The organization ID (org_ prefixed string) to bill for this transcription.

transcription_options object required

Core transcription configuration.

Show properties
  • audio_url string (required) -- HTTPS URL of the audio file to transcribe.
  • language string (required) -- Language code (e.g., en, es, fr, de). If the language is unknown, use en and auto-detection will be applied.

speaker_analysis object optional

Speaker diarization configuration.

Show properties
  • diarization_enabled boolean -- Enable speaker identification. Defaults to false.
  • speakers_expected_count integer -- Expected number of speakers (1-10).

intelligence_features object optional

Advanced AI analysis features.

Show properties
  • sentiment_analysis_enabled boolean -- Enable sentiment analysis per sentence. Defaults to false.
  • entity_detection_enabled boolean -- Enable named entity detection. Defaults to false.
  • auto_highlights_enabled boolean -- Enable automatic key phrase highlights. Defaults to false.
  • summarization_enabled boolean -- Enable automatic summarization. Defaults to false.

privacy_settings object optional

Privacy and PII redaction settings.

Show properties
  • pii_redaction_enabled boolean -- Enable PII redaction in transcripts. Defaults to false.
  • pii_policies string[] -- List of PII policies to apply (e.g., ["email", "phone_number", "ssn"]).

webhook object optional

Webhook notification configuration.

Show properties
  • url string -- HTTPS URL to receive webhook notifications.
  • events string[] -- List of events to trigger the webhook (e.g., ["transcription.completed", "transcription.failed"]).

Returns

Returns a success envelope with the following top-level fields:

  • success -- true on successful creation.
  • data -- Transcription data including transcription_id (with trans_ prefix), initial status of queued, model_key, audio_metadata, and timestamps.
  • billing -- Billing details including synapses_consumed, synapses_refunded, transaction_id, currency, and applied multipliers (base, organization, priority).
  • metadata -- Request metadata including request_id, timestamp, and processing_time_ms.
  • actions -- HATEOAS-style links to related endpoints such as check_status.
cURL
curl -X POST https://api.aitronos.com/v1/audio/transcribe \
  -H "Authorization: Bearer $ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "organization_id": "org_abc123def456",
    "transcription_options": {
      "language": "en",
      "audio_url": "https://storage.example.com/meeting-recording.mp3"
    },
    "speaker_analysis": {
      "diarization_enabled": true,
      "speakers_expected_count": 2
    },
    "intelligence_features": {
      "sentiment_analysis_enabled": true,
      "entity_detection_enabled": true,
      "summarization_enabled": true
    },
    "privacy_settings": {
      "pii_redaction_enabled": true,
      "pii_policies": ["email", "phone_number", "ssn"]
    }
  }'
Previous page
Next page