OrukZero

Detect human or AI speech with OrukZero

English audio, 1–60 seconds, WAV/FLAC/OGG/MP3, at most 12 MiB. Uncalibrated scores with a fixed 0.5 threshold. A valid Oruk API key belonging to an active account is required; no paid subscription is needed; OrukZero is included at no extra charge and consumes no credits or plan minutes. Long-audio aggregation is experimental.

post/v1/audio/synthetic-speech

Request

  • Base URL: https://speech-api.oruk.ai
  • URL: https://speech-api.oruk.ai/v1/audio/synthetic-speech
  • Auth: HTTP bearer

Response

Successful Response

object'synthetic_speech.result' required
model'orukzero' required
model_revision'2026-10-05-adapt-216' required
durationnumber required
prediction'human' | 'synthetic' required
synthetic_scorenumber required
threshold0.5 required
score_type'uncalibrated_sigmoid' required
aggregation'coverage_weighted_mean_logit' required
warningsstring[] required
idstring required
usageobject required

Changes

No recorded changes to this operation across all 1 revision of this API.