Text-to-speech that just works
in every language you ship.
One simple API for natural, multilingual speech. Pay per audio minute, not per seat. Set it up in a minute, whether you're a developer, a small business, or an AI agent.
No credit card. Free minutes to start.
Quickstart
Your first audio in under a minute
One request in, an audio file out. Same shape as the OpenAI speech endpoint, so most clients already know how to talk to it.
curl https://api.dinoco.ai/v1/audio/speech \
-H "Authorization: Bearer $DINOCO_API_KEY" \
-d '{
"input": "Your order is on its way.",
"voice": "mira",
"language": "en"
}' --output hello.mp3Give your agent a voice with one tool
Drop this tool definition into any agent framework. The agent calls it, Dinoco returns audio. No SDK to learn, no auth dance.
{
"name": "speak",
"description": "Convert text to natural speech",
"parameters": {
"text": { "type": "string" },
"language": { "type": "string" },
"voice": { "type": "string" }
},
"endpoint": "POST https://api.dinoco.ai/v1/audio/speech"
}Why Dinoco
Simple, affordable, built for volume
Everything you need to put speech in your product. Nothing you have to pay for but never use.
Speak your users' language
Natural voices in many languages, each one tested before we ship it. Change one field to switch language and keep the same voice.
Pay for audio, not seats
Billed per generated minute. No plans, no monthly minimum, no per-user pricing. Ideal if you're small now and big later.
Agent-ready out of the box
A plain HTTP endpoint with a clear schema. Agents, workflows, and no-code tools call it like any other tool.
Thousands of lines, one call
Need a whole catalog voiced? Send a batch, get every file plus a manifest back. High volume without babysitting.
Consistent, product-grade audio
Same loudness, format, and voice across every file. Your app sounds like one product, not a hundred clips.
Simple enough to skip the docs
Three fields. One endpoint. If you've used any speech API before, you already know this one.
How it works
Three steps. That's the whole thing.
No pipeline to build. Write, pick a voice, play.
Write your text
Any length, any supported language. Plain text in, nothing to pre-process.
Pick a voice
Choose a voice and language. Preview it in the playground before you commit.
Get your audio
A file comes back in seconds. Stream it, store it, or hand it to your agent.
Who it's for
One API, three ways to ship
Same endpoint, same per-minute price. Your code, your marketing, or your agent can do the talking.
Developers
Add voice to an app in an afternoon. Same shape as the OpenAI endpoint, so your existing client works.
- Voice for notifications & onboarding
- Read-aloud for articles and docs
- Localized audio in every market
Small businesses
Product videos, phone menus, ads, training. Voiced in-house for a fraction of what a studio charges.
- Product and explainer videos
- Phone menus and voicemail
- Social clips in multiple languages
AI agents
Give assistants and workflows a voice with one tool call. Works with any framework that can hit an HTTP endpoint.
- Voice replies for chat agents
- Automated call and support flows
- Narration for generated content
Pricing
One price. Per minute of audio.
No tiers to pick, no seats to count. You pay for the audio you generate, and that's it.
Per minute
of generated audio
Never billed
- Failed requests
- Seats or team members
- Idle time
- Same rate in every language
- Free minutes on every new account
- No monthly minimum, cancel anytime
- Volume rates for high-usage accounts
Exact per-minute rates are shared with your invite during beta.
FAQ
Questions, answered
Something else on your mind?
Talk to usHear it for yourself. Free minutes included.
Join the beta, grab your API key, and have your first audio file in under a minute.
- No credit card
- Free minutes on signup
- Cancel anytime
One voice, three languages
Your order is on its way.
ENTu pedido está en camino.
ESआपका ऑर्डर रास्ते में है।
HIIllustrative. Real samples ship with your invite.