Create Your Voice. Get an API.

Short answer: Speakvora is a custom voice API. Record about 20 seconds, or describe the voice you want, and you get a voice id and an API key you can call from any app. Your own voice is $6 per million characters, with 100,000 characters free and no subscription.

Create your voicePricingDocsPlayground

How it works

  1. Create your voice. Voice cloning: record in the portal or upload a clear file. Voice design: describe it and tune it. You hear a sample before you save.
  2. Get your API. The portal gives you a voice id and an API key straight away.
  3. Call it. One request returns an audio link. Use it from your server, phone system or app.
curl -X POST https://speakvora.com/api/v1/speak \
  -H "x-api-key: $SPEAKVORA_KEY" \
  -H "content-type: application/json" \
  -d '{"text":"Thanks for calling. How can I help?","voice":"YOUR_VOICE_ID"}'

Voice cloning API or voice design API: pick either

Clone

Your own voice

About 20 seconds of speech, or an uploaded file of 10 to 60 seconds. You confirm it is your voice or you have the speaker's permission.

Design

Describe a voice

Say what it should sound like and adjust age, pitch, pace and warmth. No recording needed.

What you get

  • A private voice only your account can use, with a voice id for the API.
  • Ready-made starter phrases prepared in your voice, so common lines are instant.
  • An API key, usage and billing in one portal, and a playground to try requests.
  • An MCP server and SDKs for JavaScript and Python if you build with AI assistants.

Price

Your own voice: $6 per million characters. Catalog voices: $0.70 per million. For reference, Google Cloud lists its Instant custom voice at $60 per million characters (official pricing, checked 2026-10-08; a different product, so this compares cost, not quality). Full numbers are on the pricing page.

Limits we state up front

  • Custom voices speak English today; French and German are beta.
  • A brand-new sentence is generated live and can take seconds to minutes; repeats are instant.
  • Imitating a specific real person by description is not allowed.

Questions

What is a custom voice API?

It lets your software speak in a voice you made, by calling one endpoint with your text. With Speakvora you create the voice in the portal, get a voice id and an API key, and send POST /v1/speak.

Is this a voice cloning API?

Yes. Record about 20 seconds of your own voice, or upload a clear 10 to 60 second file, confirm it is your voice or you have permission, and we build a voice you can call from the API. Custom voices speak English today; French and German are in beta.

What is the voice design API?

If you prefer not to use a real voice, describe it (age, pitch, pace, warmth) and tune it with the controls, listen to a sample, then save it as a voice you can call. Designing a voice that imitates a specific real person is not allowed.

How much does it cost?

Your own voice is $6 per million characters, billed per character after your free 100,000 characters. Catalog voices are $0.70 per million. There is no subscription or seat fee.

How fast is it?

Sentences that are already in our library or have been spoken before answer in a fraction of a second. A brand-new sentence in a custom voice is generated live and can take from several seconds to a few minutes depending on load.

Is my voice private?

Your custom voice and the audio made with it are private to your account. Audio is kept for 75 days after its last use and you can delete a voice, a recording or everything from the portal.

Open Voice StudioPricingDocsPlayground