Frequently Asked Questions

Heirloom is a private memoir-capture service for families. It guides a parent, grandparent, or any loved one through 100+ life-story questions — by voice or text — then organizes the answers into a searchable archive. Optionally, we turn the answers into a printed hardcover book, a digital memoir, and (with consent) a voice-enabled interactive legacy that descendants can ask questions of.

Pick a time (Sunday afternoon is most common). The subject answers 3–5 questions, by speaking or typing, in their preferred language. Sessions average 18 minutes. Over 20–30 sessions, you'll have covered their whole life.

Most of our subjects aren't. Three paths:

• You interview them (by phone/video) and record — this is how most families do it.
• Text-only with you helping them type.
• AI 引导式访谈 — Heirloom 逐一提出问题，长辈只需开口讲述即可。

The interface runs in 9 languages: English, Korean, Japanese, Chinese (Simplified), Spanish (neutral Latin American), Portuguese, French, German, Tagalog. The question bank is professionally translated for cultural nuance — we don't use raw machine translation for the questions.

Keepsake: $199 一次付清 — 1 位讲述者、引导式访谈，以及全家珍藏的记忆之书（可打印 + PDF）。无需订阅。
Family: 每年 $149 — 包含 Keepsake 的全部功能，外加家庭共享和随时可对话的家庭议会 AI。
Heritage: 每年 $299 或一次付清 $499 — 在此基础上加入讲述者本人的真实声音克隆。
查看方案并开始 →

Yes. Cancellations take effect at the end of the current billing period. Your content stays accessible for export for 30 days after cancellation. Full refund within 30 days of first payment — no questions asked.

You get 30 days to export (PDF, audio, raw data). After 30 days, memoir content is permanently deleted. Billing records are retained 7 years for tax compliance. See Privacy Policy.

The first 3 questions are free — no credit card. You can also get a full refund in the first 30 days. We'd rather people cancel than get stuck with something they don't love.

Yes — with their explicit, recorded consent. Voice cloning requires a separate, two-step consent flow: the subject records a consent phrase, and can revoke it at any time. See Consent Policy.

Quality scales with input. Most subjects produce a recognizable voice clone after ~2 hours of clean recordings. For a truly natural clone, we recommend 6+ hours spread over multiple sessions. Quality varies with accent, age, health, and recording environment. We don't make promises until we hear the voice.

We explicitly limit AI responses to content derived from captured answers. When the AI doesn't know, it says so — it does not fabricate. Your dashboard shows the source answer for every generated response.

AI processes your content per-request, under contracts with vendors (Anthropic, Deepgram) who are prohibited from retaining or training on it. Your content is never used to train public models. Ever.

Yes. Informed consent is required. You can capture verbal consent at the start of the first session — we provide the script in your dashboard. Never record someone who doesn't know they're being recorded.

We pause AI generation and external sharing, offer a 30-day mediation window, and defer to the original subject's documented wishes (or the purchasing account holder's decision if no wishes are documented). Full details in Consent Policy.

The account holder retains access per their original configuration. Voice cloning remains available only if the subject consented during life. On request from verified next of kin, we transfer ownership or honor deletion. We respect the original consent, always.

由讲述者的回答汇编而成的记忆之书（可打印 + PDF）。所有方案均包含，全家人可终身阅读。

包含 Family 的全部功能，外加讲述者本人的真实声音克隆（ElevenLabs，需录音同意）。从未见过他们的后代提问时，也能听到他们本人的声音回答。每年 $299，或一次付清 $499。开始 →

Yes. Our Creator Program pays 25–50% recurring commission for 12 months on every paid referral.

Yes. Our Partnership Program is designed for estate attorneys, hospice teams, funeral homes, senior living operators, and faith communities. White-label options available at the organizational tier.