- Voice memory
- A voice recording preserved for the purpose of remembering who someone was — typically a story, message, or reflection captured in their own voice. The core concept behind LifeEcho's product category.
- Voice memory service
- A product designed specifically for capturing and preserving voice recordings of loved ones, typically including recording, transcription, storage, and sharing features. Distinct from general transcription tools (like Rev) or meeting transcribers (like Otter.ai) in that it is purpose-built for families and long-term preservation.
- Oral history
- The practice of collecting spoken recollections from individuals about historical events, family stories, or cultural traditions. Distinct from written history in that it preserves the actual voice and speech patterns of the source. LifeEcho's phone-call workflow makes it a practical way to build oral history archives for families.
- AI transcription
- The automatic conversion of spoken audio into written text using artificial intelligence. Modern AI transcription handles a wide range of accents and speaking styles with high accuracy. See how AI transcription works for family phone calls for a plain-English walkthrough.
- OpenAI Whisper
- An open-source speech recognition model released by OpenAI in 2022, trained on ~680,000 hours of multilingual audio. Currently one of the most accurate general-purpose speech-to-text systems. LifeEcho uses Whisper for all transcription.
- AI voice cloning
- The generation of synthetic speech in a specific person's voice, trained on audio samples of that person. Can produce speech the person never actually said. See AI voice cloning vs. real voice recordings: honest truth for an ethical discussion of this technology in family contexts.
- Voice avatar
- A conversational AI system trained on a person's voice and biographical information, designed to allow others to have simulated interactions with the person — often used after the person has died. The core product of UK-based Life's Echo; see LifeEcho vs Life's Echo for a full comparison.
- Griefbot
- A chatbot or AI system designed to simulate communication with a deceased person, often using their recorded voice, writings, or social media history as training data. Sometimes called "deathbots." See griefbot AI vs. real voice recordings.
- Ethical will
- A document or recording in which a person shares their values, life lessons, and messages for future generations — distinct from a legal will in that it conveys meaning rather than property. Voice recordings make particularly resonant ethical wills.
- Voice memoir
- A memoir in voice form: a structured collection of recorded memories, stories, and reflections that together tell the arc of a person's life. Can later be converted to a written document via AI memoir export.
- AI memoir
- A chapter-organized written memoir generated automatically from voice recordings. The AI reads transcripts, groups related stories into chapters, and produces a book-length document in the speaker's own words. See what an AI memoir looks like for a concrete example.
- Semantic search
- Search that matches on meaning rather than exact words. Allows finding "the story about Dad's first car" even when the search result's transcript doesn't contain the word "car." See semantic search for family memories.
- Embedding
- A numerical representation of text that captures its meaning. Used in semantic search and retrieval-augmented AI to compare texts by meaning rather than exact wording. Every AI-powered semantic search uses embeddings under the hood.
- Word-level timestamps
- Transcription output where every word carries a precise timing marker indicating when it was spoken. Enables click-to-audio navigation and precise quotation from recordings. LifeEcho produces word-level timestamps on every recording.
- Speaker diarization
- The process of automatically identifying and labeling distinct speakers within a multi-speaker recording. Often needed for meeting transcripts; less common in single-speaker voice memoirs. Currently not supported in LifeEcho (single-speaker focus).
- Retrieval-augmented generation (RAG)
- An AI technique that retrieves real source material (e.g., actual transcripts from family recordings) and uses it to ground AI responses in verified information. Used for Q&A over personal memories without the AI hallucinating or inventing content.
- Voice print
- A unique acoustic signature of a person's voice, analogous to a fingerprint. Used for speaker identification and, increasingly, for voice cloning training data. Raises both biometric privacy and post-mortem consent questions.
- Deepfake
- Synthetic media — audio, video, or images — generated by AI to convincingly imitate a real person. Voice deepfakes are increasingly realistic and raise ethical and authenticity concerns for memorial and impersonation use cases.
- Digital legacy
- The collection of digital assets a person leaves behind — accounts, files, messages, recordings — and the plan for how those assets are preserved, accessed, and used after their death. Voice memory libraries are increasingly treated as a core part of digital legacy planning.
- Audio archive
- A collection of preserved audio recordings, typically organized, searchable, and backed up for long-term access. A family voice memory library is a specific kind of audio archive, structured around personal and emotional value rather than institutional or broadcast use.
- Voice journaling
- The practice of regularly recording spoken reflections — similar to writing in a diary, but in voice. Can be done by the person themselves or through prompted conversations with family.
- Voicemail preservation
- The practice of saving and backing up voicemail recordings, particularly from deceased loved ones, before phone carriers automatically delete them. Often one of the first places families look for voice recordings after a loss.
- Personal history recording
- A recording made specifically to document the life story of an individual — typically longer-form, possibly conducted across multiple sessions, and often guided by structured questions like the ones in LifeEcho's 50-question life story guide.
- Life story capture
- The broader practice of preserving a person's complete life narrative through any combination of voice recordings, writings, photos, and interviews. A core use case LifeEcho is designed to support.
- Memory prompt
- A structured question designed to draw out a meaningful story or reflection. Examples: "What was your childhood home like?" or "What advice would you give your younger self?" Used to make recording sessions more productive than open-ended conversation. LifeEcho includes a curated library of memory prompts.
- First-person summary
- A summary of a voice recording written in first-person perspective (as "I") rather than as a third-person report about them. Produces summaries that feel like personal notes rather than surveillance logs. LifeEcho generates first-person summaries automatically via GPT on every recording.
- Voice memory book
- A collection of voice recordings organized as a keepsake — the audio equivalent of a written memory book. Can be presented as a digital library or converted to a printable memoir via AI memoir export.
- Auto-tagging
- Automatic categorization of recordings by theme (e.g., childhood, family, work, faith). Typically done via AI that reads transcripts and assigns relevant tags without human input. On the LifeEcho roadmap.
- Speakable content
- Content formatted for voice assistants to read aloud. In schema.org, the SpeakableSpecification identifies which parts of a page are suitable for text-to-speech reading by Google Assistant, Alexa, and similar.
- Lifetime storage
- A storage guarantee that recordings will be preserved for the lifetime of the account, typically decades, with no risk of automatic deletion due to inactivity. Critical for family memory libraries that need to outlive the person who created them. Included on LifeEcho's paid plans.
Ready to start a voice memory library of your own?
Free 15-minute trial. Any phone. No credit card.
Start Your First Recording