Lalia for Business

Give every book in your library a voice — in any language.

The API behind Lalia. Add AI audiobook narration, page-level translation, and a 75,000-title public-domain catalog to your app, e-library, or learning platform. One REST call per page. Pay only for what your users play.

Get your API key Read the docs

No credit card required · first 1,000 calls free · live in minutes

narrate-a-page.sh

curl https://api.laliabook.com/v1/narrate \
  -H "Authorization: Bearer lalia_sk_live_…" \
  -d '{
    "text":     "It was the best of times…",
    "language": "es",
    "voice":    "amber"
  }'

# → 200 OK in ~1s
{
  "status":    "ready",
  "audio_url": "https://api.laliabook.com/v1/audio/narr_9f2c…",
  "voice":     "amber",
  "language":  "es"
}

Built for teams like

e-libraries·ed-tech platforms·publishers·language schools·accessibility tools

Why Lalia

Everything readers expect, one API away

The same engine that powers the Lalia app — battle-tested narration, translation, and caching — exposed as clean REST endpoints your team can ship this week.

🔊

Studio-quality narration

Natural Gemini-TTS voices narrate any page of text in 10+ languages. Five distinct voices, tuned for long-form listening.

🌍

Read in any language

Translate pages on the fly, then narrate the translation. Spanish, French, Somali, Arabic, Chinese and more — same call, one parameter.

📚

75k-title catalog included

Search and import from a curated public-domain catalog with covers and summaries — stock your library without licensing a single title.

⚡

Cache-first pricing

Popular pages are narrated once and served from cache — repeat plays are fast and you're never billed twice for the same audio.

Integration

Live in three steps

Designed so one engineer can integrate it in an afternoon — no SDKs required, just HTTPS and JSON.

Create a key

lalia_sk_live_e91b4c…

Call the API

Send a page of text to /v1/narrate — get back a streaming audio URL in about a second.

POST /v1/narrate → { "audio_url": … }

Press play

Drop the URL into any <audio> element or player. Usage is metered per call and billed monthly.

<audio src={audio_url} controls />

Pricing

Pay per call. Nothing up front.

Metered like a utility: every request has a listed price, cached repeats are free, and volume discounts kick in automatically.

Developer

Prototype and launch

$0 to start

First 1,000 calls free every month
Then standard per-call rates
All endpoints, all voices
Community support

Start free

Growth

Libraries & ed-tech in production

−25% per-call rates over 100k calls/mo

Volume discounts applied automatically
99.9% uptime SLA
Shared narration cache — repeat plays free
Email support, 24h response

Get your API key

Enterprise

Publishers & institutions

Custom annual agreement

Committed-use pricing
Custom voices & private catalog ingestion
SSO, audit logs, DPA
Dedicated support engineer

Talk to us

Standard per-call rates

Endpoint	What it does	Price per call
POST /v1/narrate	Narrate one page (up to 6,000 chars) to audio	$0.012
POST /v1/translate	Translate one page into a target language	$0.004
GET /v1/catalog/search	Search the 75k-title public-domain catalog	$0.0005
GET /v1/usage	Your metered usage & spend, for your own dashboards	Free

Do the math: narrate a 300-page book for about $3.60 — once. Every reader after that streams it from cache, free.

~$3.60

One-time cost to give a full 300-page book a voice with the Lalia API

$1,000+

Typical studio cost for the same book — human audiobook production runs $100–$400 per finished hour

Cost of every repeat listen — cached narrations are never billed twice

FAQ

Common questions

Can we narrate our own copyrighted content?

Yes. You send the text, we return the audio — your content never enters the shared catalog and is only cached for your account. You must hold the rights to any text you submit.

How is billing calculated?

Every billable response includes its unit price. Calls are summed daily, volume discounts apply automatically past 100k calls/month, and you receive one monthly invoice. Cached repeat narrations are free.

What about latency and uptime?

Cached pages return in under 300 ms; fresh narration typically completes in 1–3 seconds per page. Growth and Enterprise plans include a 99.9% uptime SLA.

Do you offer white-labeling?

The API is fully white-label: no Lalia branding appears in responses or audio. Enterprise plans can add custom voices tuned to your brand.

Your library, listening-ready by Friday.

Create a key, make your first call, and hear it speak — all before lunch.

Get your API key — free