How to get started with cloning my voice?
You can get started on our self-serve platform in a few simple steps. Create an account, click on Build a Voice and start recording the sentences that pop up.
How much data is needed?
We need a minimum of 50 sentences recorded to kick off the training, but the more data you record the better the quality of the cloned voice. Training is done in increments of 50 sentences. For examples after the 50 sentences have been recorded, we will train and deploy your voice. To improve the voice, the next training will be done after you have recorded 50 or 100 more sentences.
Can I use Resemble to clone somebody else’s voice?
Yes, but only if you have the consent from the third party and the third party is aware of the use case for their synthetic voice. Our Professional plan allows you to upload data but it’s still subject to approval. Please read more about it on our Ethics page.
Can I fine-tune or apply emotions to the audio?
Yes, with our editor, you will be able to fine-tune it to your liking. The Editor makes it super simple for anyone to apply granular control on the audio. We will also be adding the ability to apply different emotions in the near future.
How can I create content in my cloned voice?
You will be notified via email once your voice has been cloned and is ready to use. You can consume it via our web platform or via API.
Can we license a voice from Resemble for our brand?
Yes. We offer ficticious voices that could be licensed from us. You can license the voices for a year or however long you want it to be the voice of your brand. Read more about our ficticious voices here: https://www.resemble.ai/fictitious-2/