Audio
Text-to-speech, transcriptions, and realtime session minting
OpenAI-compatible audio surfaces on Hypery. All three endpoints require OAuth scope ai:audio (or a first-party API key / session). Chat scopes alone are not accepted.
Billing
Audio is not token-metered the same way as chat. Hypery always charges the billed team:
Also enforced on every call: DISABLE_AI_SPEND, rate limits, app + team spending caps, model permission spend caps, usageRecords.
Realtime mint charges a floor ($0.50). Ongoing WebRTC media is terminated by the upstream — clients must settle remaining cost via POST /api/v1/usage/charge with category=realtime and scope billing:charge when the session ends.
Text-to-speech
POST /api/v1/audio/speech
Optional header: Idempotency-Key (prevents double-charge on retries).
Transcriptions
POST /api/v1/audio/transcriptions — multipart/form-data
Billing estimates minutes from file size (~32 kbps speech heuristic, capped at 45 min) × $0.006/min.
Realtime sessions
POST /api/v1/realtime/sessions — mints an ephemeral client_secret.
Response (whitelisted):
{
"id": "…",
"model": "gpt-4o-realtime-preview",
"client_secret": { "value": "…", "expires_at": 0 },
"hypery": {
"sdp_url": "https://…/v1/realtime",
"settle_usage": {
"path": "/api/v1/usage/charge",
"category": "realtime"
}
}
}Accepted body fields include standup options: input_audio_transcription, turn_detection.