$5 per 1M characters on Pro and Max

Text to Speech API for $5 per Million Characters

Realistic MP3 speech from one REST call: 500+ voices in 150+ languages, voice cloning and voice design. On Pro and Max, every extra million characters costs just $5, roughly 15 hours of narration.

 

Start free with 10,000 credits a month. No credit card needed.

Get your free API key
$5per extra 1M characters
10,000free credits a month
500+voices

API pricing

$5 per million characters

One credit is one character, shared with Lazybird Studio. Pick the plan that covers your monthly volume, then top up at $5 per 1M on Pro and Max. Extra credits never expire.

Pro · $19/month

2M characters a month, then $5 per 1M

Need 5M characters this month? That's $19 + 3 × $5 = $34. Commercial rights included.

See Pro

Max · $39/month

10M characters a month, just $3.90 per 1M

Built for high-volume apps. Go past 10M and every extra million is still $5. Commercial rights included.

See Max

All API plans

API usage shares the same monthly credits as Lazybird Studio. One credit is one character.

Plan Price Credits per month Extra credits per 1M
Free $0 10,000 $30
Creator $9/month 500,000 $20
Pro $19/month 2,000,000 $5
Max $39/month 10,000,000 $5

Extra credits never expire and you can buy any amount from $10. Paid plans include commercial usage rights.

Text to Speech API for Developers

The Lazybird Text to Speech API gives your app the same voices as Lazybird Studio through a simple REST API. Send text, get back an MP3.

  • 500+ voices, 150+ languages: multilingual voices can switch language with a single setting.
  • Voice cloning: create a private voice from a short, clean recording and use it anywhere a voice ID works.
  • Voice design: describe a speaker ("a calm, older British narrator") and get a brand-new voice to audition and keep.
  • Fine control: adjust speed, pitch and volume, and pick speaking styles on voices that support them.
  • Phone-ready audio: ask for an 8 kHz WAV (PCM or µ-law) for voicemail greetings and phone menus instead of an MP3.

One request, one MP3

Authenticate with your key in the X-API-Key header and post your text:

curl -X POST "https://api.lazybird.app/v2/generate-speech" \
  -H "X-API-Key: $LAZYBIRD_API_KEY" \
  -H "Content-Type: application/json" \
  -o speech.mp3 \
  --data '{"voiceId": "tgJ2Hes6DauwTg8kFrRPLN", "text": "Hello from Lazybird!"}'

That voice ID is Elyssia, our default voice. Find others with GET /v2/voices?q=narrator, check your balance with GET /v2/me, and find every endpoint in the API reference and the OpenAPI spec.

SSML Support

Plain text is enough for most scripts. When you need more control, send SSML in the ssml field instead:

  • Pauses: <break time="700ms"/> adds a precise pause, up to 60 seconds.
  • Pronunciation: <phoneme alphabet="ipa" ph="..."> fixes names, brands and technical terms.

Built for agents too

Building with AI? Lazybird's MCP server lets ChatGPT, Claude and other agents generate speech, and our AI agents guide shows how to set it up in a few clicks.

Get your API key and start building with the best text to speech API today.

Get Started

Many vibes, many voices

Ditch the monotonous voice-over with hundreds of characters, accents and tones to choose from. Youthful or mature, fast or slow, always find the perfect voice for your content.
Ava
Ava
American English
Davis
Davis
American English
Andrew
Andrew
American
Agnieszka
Agnieszka
Polish
Henri
Henri
French
Laia
Laia
Spanish
Nanami
Nanami
Japanese
Oliver
Oliver
British English
Amala
Amala
German
Try Now

Frequently asked questions for API