Frequently asked questions.

The first time you hear it, sometimes a little. Then you hear your dad say your daughter's name and you remember why you started. We've thought a lot about the line between "preserving a voice" and "pretending someone is still here." This is the first one. We won't ever pretend to be the second.

Quality depends on how much clean audio you provide. With 5 minutes of clear recordings, the result usually sounds recognizable. With 30+ minutes, it sounds genuinely like the person, including small mannerisms and rhythm.

Conversational chat is coming. For now, you can type any text and hear it spoken back in their voice, save audio messages, and share clips with family.

Parents and legal guardians can create profiles for kids. We treat children's data with extra care and never market to anyone under 18.

Yes — but they have to consent. We email them, they record a short consent statement, then we make the profile. There's no workaround, on purpose.

Clean speech in a quiet room. Voice memos and voicemails work great. Old videos work too — we extract the audio. Background noise, music, and multiple people talking at once are harder for us to use.

For a basic profile, sometimes — even thirty seconds can produce something recognizable. For a higher-fidelity profile, we'd suggest five minutes or more if you can find it. Our Audio Coach will be honest with you about what's possible from what you have.

We have a process for that, and we use it. Profiles can be paused while families talk through concerns. We don't take sides, and we don't override family wishes.

That's okay. You can save your account and come back. Many people find the first weeks too soon, and revisit months or years later. There's no deadline, on us.

We never sell your audio, your voice profile, or any data about you. Voice profiles are encrypted with per-account keys. We share data only with the technical providers required to make the product work — see our Privacy Policy for the list. You can export or delete everything, any time.

Every output carries an inaudible watermark. We require explicit consent for non-self voices. We refuse to clone public figures, politicians, or anyone you don't have a personal relationship with. We respond to misuse reports within 24 hours.

Ninety days of advance notice. A full export of everything you've stored. Help getting it somewhere safe. It's in our terms, not just on this page.

Email report@heirloomvocals.com or use the form at /report. Urgent reports get acknowledged within 4 hours during US business hours.

Seven days, one voice profile, watermarked playback, 50 characters of TTS per day. No credit card needed. After day 8, your vault is archived for 30 days in case you want to come back.

Yes. One click in your account settings. Your data is preserved for 30 days post-cancellation in case you change your mind, then permanently deleted.

The one-time fee anchors the seriousness of a perpetual commitment. The annual maintenance funds the actual cost of keeping the vault live forever — including the operational obligations of our shutdown promise.