ElevenLabs - Mavera

ElevenLabs’ voice AI platform — text-to-speech, voice cloning, dubbing, and voice design — pairs with Mavera to create audio-first content pipelines. These five jobs chain ElevenLabs endpoints with Mavera surfaces to clone spokesperson voices for branded audio, localize scripts into multiple languages, convert focus group findings into audio executive briefings, assign unique voices to Speak session personas, and run full ad-script-to-broadcast loops with quality feedback.

API Reference Card

Detail	Value
Base URL	`https://api.elevenlabs.io/v1`
Auth	`xi-api-key` header — `xi-api-key: your_key_here`
Key endpoints	POST /text-to-speech/, POST /voices/add, POST /v1/dubbing, GET /voices
Rate limits	Plan-based — Free: limited, Starter: moderate, Pro/Scale: high (see ElevenLabs pricing)
Mavera base	`https://app.mavera.io/api/v1`
Mavera auth	`Authorization: Bearer mvra_live_...`

All examples use two environment variables: ELEVENLABS_API_KEY (your ElevenLabs API key) and MAVERA_API_KEY (your Mavera key starting with mvra_live_). ElevenLabs uses the xi-api-key header — not Authorization: Bearer. Never commit either key to version control. Use a .env file or your platform’s secret manager.

Prerequisites

ElevenLabs API key

Sign up at elevenlabs.io. Navigate to Profile → API Keys and copy your key. Ensure your plan has sufficient character credits.

Mavera API key

Get your key from Mavera dashboard.

Install dependencies

# Python
pip install requests

# JavaScript (Node 18+ has built-in fetch)
npm install node-fetch

Set environment variables

export ELEVENLABS_API_KEY="your_elevenlabs_key"
export MAVERA_API_KEY="mvra_live_xxxxx"

Jobs

#	Job	ElevenLabs Endpoint	Mavera Surface	Output
1	Persona Voice Cloning → Branded Audio	POST /voices/add, POST /text-to-speech/	Generate	Branded audio in cloned voice
2	Multilingual Content Localization	POST /v1/dubbing	Generate	Localized audio in multiple languages
3	Focus Group Results → Audio Report	POST /text-to-speech/	Focus Groups, Mave Agent	Audio executive briefing
4	Speak Session Enhancement	POST /text-to-speech/	Personas, Speak	Multi-voice audio output
5	Ad Script → Production-Ready Audio	POST /text-to-speech/	Generate, Mave Agent	Broadcast-ready audio with quality loop

Rate Limits & Production Notes

Plan	Characters/month	Concurrent Requests	Strategy
Free	10,000	2	Development and testing only
Starter	30,000	3	Single-campaign workflows
Creator	100,000	5	Multi-variation production
Pro	500,000	10	Batch pipelines with retry logic
Scale	2,000,000	15	Full localization + multi-voice sessions

Use eleven_turbo_v2_5 for faster, lower-cost TTS during development and iteration. Switch to eleven_multilingual_v2 for final production renders. Turbo models return audio 2-3x faster — ideal for the quality feedback loop in Job 5.

ElevenLabs uses xi-api-key as the authentication header — not Authorization: Bearer. Every request must include "xi-api-key": "your_key" in headers. Character limits reset monthly. If you receive a 401 with quota_exceeded, your monthly allocation is exhausted — upgrade or wait for the next billing cycle. Binary audio responses must be written to files, not parsed as JSON. Monitor character usage at ElevenLabs dashboard. Monitor Mavera credits at Dashboard.

​API Reference Card

​Prerequisites

​Jobs

​Rate Limits & Production Notes

All Integrations

ElevenLabs API Docs

Generate

Focus Groups

Mave Agent

Brand Voice

API Reference Card

Prerequisites

Jobs

Rate Limits & Production Notes