Commission an AI voice

Learn how to create a custom AI voice for your company to use.

If you are interested in creating a custom AI voice for your company to use, BeyondWords can help. We have worked with various voice artists and publishers to produce AI voices in multiple languages.

Plus, our all-in-one audio CMS allows you to use custom voices to their full potential.

Register your interest

How do I commission an AI voice?

In order to commission an AI voice, you will need to recruit a speaker whose voice you wish to clone. This could be a voice actor or a writer on your team, for example. Please ask if this is something you need help with.

You will then need to form a Voice Services Agreement with the speaker. This is the legal contract that sets out the terms and conditions for both parties. Our Voice Services Agreement Template (VSAT) provides a good starting point — we recommend working with an independent legal advisor to tailor the contract to your needs.

You will also need to form a contract with BeyondWords.

Once the terms have been agreed, the voice recordings should be made as per the Voice Services Agreement. We will provide a script, which contains 2,000–8,000 utterances (2–8 recorded hours). This script has been designed to ensure the voice model is trained optimally.

These recordings will then be passed to our text-to-speech research team for pre-processing. This involves removing errors and silences, converting to the appropriate format, and computationally extracting the thousands of characteristics that make up the speaker's voice.

This speech data is then used to train a voice model, which goes through objective and subjective testing. Once you are happy with the result, the AI voice will be deployed and available through your BeyondWords account.

What should be included in my contract with the voice provider?

We recommend using our Voice Services Agreement Template (VSAT) and seeking independent legal advice when commissioning an AI voice.

Produced in collaboration with the Open Voice Network, the VSAT facilitates fair contracting between a person recording their voice for the purposes of synthetic speech and the company using the recordings. It covers areas a range of commercial and contractual areas, and is free to use.

Are your voice clones high quality?

BeyondWords is committed to developing the highest-quality voice clones.

In this video, you can hear voice actor Joe Coen's voice compared against his AI voice. Visit our blog to learn more about the creation of Joe's AI voice.

If you are interested in commissioning a custom AI voice, email [email protected].


Did this page help you?