TTS API async jobs and webhooks for long scripts — Audra
Audra is an OpenAI-compatible batch TTS API at $25/1M characters on Essential, with −16 LUFS mastering and QA-gated billing. Not for realtime voice agents or cloning.
Who this is for
- Teams rendering long-form batch narration (courses, podcasts, docs, cron)
- Pipelines on the OpenAI SDK or plain HTTP
- Builders who want prepaid $/char without per-seat tax
- Anyone focused on jobs + webhook completion
What you get on Essential
Essential ships 72 voices across 10 languages, −16 LUFS mastering, and QA-gated billing on an OpenAI-compatible API. Sync requests accept up to 50k characters; longer scripts use POST /v2/speech/jobs (up to 500k) with optional webhook_url.
- Model: audra-essential (plus / pro / conversations on /pricing#engines)
- Voice: slug from /voices
- Format: mp3 (default) or wav
- Optional language BCP-47 when the slug does not imply locale
curl https://audratalks.com/v2/speech/jobs \
-H "Authorization: Bearer sk-live-YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"audra-essential","text":"Long script…","voice":"june","format":"mp3","webhook_url":"https://example.com/hooks/audra"}'
Next: read the docs at /docs. Engine details on /pricing#engines; migration snippets on /migrate.
Cost sanity check
Want to hear a sample on your own script? Get an API key at /dashboard — 25k free characters, no card.
FAQ
How do I know a job finished?
Poll the job status endpoint, or pass webhook_url on create so Audra POSTs completion to your server.
Are failed jobs billed?
Credits reserve at start and finalize after QA. Failed gates auto-refund — see /blog/tts-api-billing-qa-gated.
Does Audra support realtime voice agents?
No. Audra is a batch TTS API for mastered MP3/WAV files. For sub-200ms conversational agents, stay on a realtime-oriented provider.
Can I clone a custom brand voice?
No. Audra ships preset voices only. Use ElevenLabs (or similar) when cloning is a hard requirement — see /vs/elevenlabs.
Next steps
- Primary next step: read the docs at /docs
- A/B voices on /compare
- Honest matrices: /vs/elevenlabs and /vs/openai-tts
- Reference: /faq
- Read next: /blog/tts-api-billing-qa-gated
Audra is not a clone of ElevenLabs realtime or voice cloning. If that is your core product, stay there. If you need batch narration economics with broadcast loudness, start at /dashboard.