Transcripts

Transcribe audio or video — estimate

Price the request without executing it. Same body as POST /v1/speech/transcripts.

post/v1/speech/transcripts/estimate

Request

  • Base URL: https://apihub.co.il
  • URL: https://apihub.co.il/v1/speech/transcripts/estimate
  • Auth: HTTP bearer

Request body

promptstring

Free-text context for universal-3-5-pro · add-on

audio_urlstring uri required

Public URL of the audio or video file (mp3, wav, m4a, mp4, mov, webm…)

punctuateboolean
redact_piiboolean

Remove personal data (names, phone numbers…) from the text · add-on

format_textboolean

Casing and number formatting

audio_end_atinteger

Stop at this millisecond

disfluenciesboolean

Keep filler words (um, uh)

language_codestring

Language of the audio, e.g. he, en, ar, ru. Omit for automatic detection.

speech_modelsstring[]

Models in order of preference. Default: universal-3-5-pro, falling back to universal-2 for languages Pro does not support (e.g. Hebrew). Billed by the model actually used.

content_safetyboolean

Sensitive-content labels · add-on

iab_categoriesboolean

Topic detection · add-on

speaker_labelsboolean

Who said what (speaker diarization) · add-on

auto_highlightsboolean

Key phrases · add-on

keyterms_promptstring[]

Names and terms to recognise correctly (brands, people, jargon) · add-on

audio_start_frominteger

Start at this millisecond (only this part is billed)

entity_detectionboolean

Detect people, places, organisations… · add-on

filter_profanityboolean

Mask profanity · add-on

speakers_expectedinteger
language_detectionboolean

Detect the language automatically (default when language_code is omitted)

sentiment_analysisboolean

Sentiment per sentence · add-on

redact_pii_policiesstring[]

Example request

{
  "language_code": "he"
}

Response

Estimate

operationstring
estimated_creditsnumber
credits_to_reservenumber
estimated_usageobject
notestring nullable
sufficientboolean

Changes

No recorded changes to this operation across all 2 revisions of this API.