Custom Voice Creation: Record, Upload, or Design

Create custom voices in Speakvora: record a paragraph, upload speech, or design with tuning meters. English supported; French and German beta. Max 3 voices per account.

Short answer. Speakvora lets you create up to 3 custom voices per account by recording a paragraph, uploading 10–60 seconds of speech, or describing and tuning a voice with design meters. English is fully supported; French and German are in beta. You must confirm the speech is yours or that you have permission to use it.

How to Create a Custom Voice

You can build a custom voice in three ways through the Speakvora dashboard. Each method creates a voice that is unique to your account and ready for use in your applications.

Record a paragraph: speak naturally into your browser or device.
Upload speech: provide 10–60 seconds of your own audio (WAV, MP3, M4A, OGG, FLAC, or WebM).
Design a voice: describe what you want and tune it with meters for gender, age, pitch, pace, warmth, and energy, then listen to a sample before saving.

Language Support

English is fully supported for all three creation methods. French and German are available in beta. Other languages—Spanish, Hindi, Chinese, and Japanese—are not offered because they were not intelligible when tested with the current voice engine.

Voice Limits and Account Rules

Each account can have up to 3 custom voices. A voice becomes "ready" only after verification: Speakvora generates a test sentence and listens to confirm the voice works as expected. Custom-voice audio is stored only after a second request. Stored audio is private and automatically deleted after 75 days without use; each reuse restarts the 75-day timer.

Consent and Permissions

When you upload or record speech, you must confirm that the audio is your own voice or that you have permission to use it. This ensures you have the right to use the voice in your applications.

Pricing

Custom voices are billed at $6 per 1 million characters when you use them to generate speech.

Delivery Styles for English

If your custom voice is in English, you can apply delivery styles to add tone to your speech. Available styles are happy, excited, sorry, calm, and serious. Styles are off by default and can be requested per request.

Using Your Custom Voice

Once your voice is ready, use it with the /v1/speak endpoint by passing your custom voice ID (starting with cv_) and the text you want to speak.

Frequently asked questions

Can I create a custom voice in languages other than English, French, or German?

No. Spanish, Hindi, Chinese, and Japanese are not offered because they were not intelligible when tested with the voice engine.

How many custom voices can I create?

You can create up to 3 custom voices per account.

What happens to my custom-voice audio if I don't use it?

Custom-voice audio is stored only after a second request and is automatically deleted after 75 days without use. Each time you reuse the voice, the 75-day timer restarts.

Can I use delivery styles with my custom voice?

Yes, if your custom voice is in English. You can apply styles like happy, excited, calm, or serious to add tone to your speech.

Do I need permission to upload someone else's voice?

Yes. When you upload or record speech, you must confirm that the audio is your own voice or that you have permission to use it.