Model catalog

Models served by the oruk Speech API

Every model below is served behind one versioned API at https://speech-api.oruk.ai. A machine-readable version of this catalog is available without authentication at https://speech-api.oruk.ai/models, and authenticated clients can query GET /v1/models and GET /v1/pricing.

Input: Prerecorded audio (WAV, FLAC, MP3, M4A, OGG, WebM)
Output: Text: transcripts, emotion labels, style labels
Language: English (API v1)
Limits: 30 MB per file, 60 minutes per file
Inference location: United States (US)
Pricing version: 2026-07-11

oruk Spectra 1

Stable

oruk-spectra-1

A unified model for high-volume emotion, speaking-style, and combined affect classification through the same API contract.

Task (USD / audio minute)	Transcript	Emotion	Style	Affect	Analysis
Billed per second	$0.0045	$0.0060	$0.0060	$0.0075	$0.0090

Served from datacenters in the United States (country code US).

oruk Spectra 2

Preview

oruk-spectra-2

A preview tier listed in the model catalog while evaluation and optimization continue. It is not yet serving inference traffic — use Spectra 1 or Resonance today.

Task (USD / audio minute)	Transcript	Emotion	Style	Affect	Analysis
Billed per second	$0.0065	$0.0080	$0.0080	$0.0100	$0.0120

Preview tier: listed in the catalog and priced, but not yet serving inference traffic — use Spectra 1 or Resonance today. Served from datacenters in the United States (country code US).

oruk Resonance

Stable

oruk-resonance

The flagship English speech model for transcripts, calibrated emotion and speaking-style labels, and segment-level tagged text in one request.

Task (USD / audio minute)	Transcript	Emotion	Style	Affect	Analysis
Billed per second	$0.0080	$0.0080	$0.0080	$0.0100	$0.0120

Served from datacenters in the United States (country code US).

Endpoints

All models are reachable through the same task endpoints; pass the model ID in the model field.

POST /v1/audio/transcriptionsEnglish transcript
POST /v1/audio/emotionsCalibrated multilabel emotions
POST /v1/audio/stylesCalibrated multilabel speaking styles
POST /v1/audio/affectEmotion and style labels
POST /v1/audio/analysisTranscript, labels, segments, and tagged text

Full request and response documentation is in the API docs, rates are on the pricing page, and accuracy comparisons follow the benchmark methodology. Data handling is covered by the privacy policy and terms of service.