Voicemod requires consent before you can clone another voice

The audio task

A streamer, gamer, or musician wants to perform in real time through a transformed voice, for a character bit, an effect, or anonymity, without the latency of processing a full recording afterward. Voicemod is a real-time voice-changing application built around a virtual microphone, marketed for gaming and streaming but also used by some musicians for character voices and effects.

What the documents show

Voicemod's product page describes more than 200 preset voices plus a custom voice-creation tool called Voicelab, and states its AI voices are trained with professional voice actors under "Fairly Trained certified models," a reference to a third-party certification scheme for consented training data. The terms of use address a different scenario: a user building their own voice with the self-service tools. That document states a user "can use your own voice or the voice of another person, provided that you have their consent or permission to use their voice and to use their name for the voice you will create." The acceptable-use policy embedded in the same terms separately prohibits content that infringes "any third-party right, including any copyright, trademark, design right, patent, trade secret, moral right, privacy right, right of publicity," and bars content that is "defamatory, false, intentionally misleading."

Rights status

The terms place the consent burden on the user for any voice built from someone else's voice, and state that users are "solely responsible for User Content," whether uploaded or created with Voicemod's own tools. The terms reviewed do not contain a specific clause naming impersonation of a real, identifiable person's voice as prohibited outright; that outcome instead follows from the general consent requirement and the third-party-rights prohibition read together, not from an explicit standalone ban.

What to check before you use it

This is an editorial checklist. Before cloning any voice other than your own in Voicelab, document the source speaker's consent in writing, since the terms make that consent a condition rather than a formality. If you are using a preset AI voice for a musical release rather than for streaming or gaming, separately check whether the "Fairly Trained" claim applies to that specific voice.

• Do you have documented consent from the specific person whose voice or name you are using to build a custom voice?
• Does the platform or distributor you plan to release through require disclosure of a synthetic or transformed vocal?
• Is the preset voice you are using one of the "Fairly Trained certified" models, or an older voice not covered by that claim?

Voicemod's own terms make consent the user's obligation for any voice cloned from a real person, which is a narrower and more actionable rule than a blanket ban on impersonation would be.