Your voice is yours.
Memo is built so that we never hear, see, or store what you record. Here is exactly how that works.
Local-first, by design
Memo stores your recordings and transcripts in your browser's local storage (IndexedDB) on your device. There is no account system, no analytics, and no server that receives your audio by default.
Browser transcription
When you transcribe with Browser Whisper, the AI model runs entirely on your device. Audio is processed locally and nothing is sent over the network. Once the model is downloaded, transcription works fully offline.
Optional OpenAI transcription
If you choose OpenAI transcription, the audio you transcribe is sent to OpenAI's API using your own API key, which you provide and which is stored only in your browser. This is entirely opt-in, per recording. OpenAI's data usage policies apply to that audio.
API keys
Your OpenAI API key is stored locally in your browser and is only ever sent to OpenAI's API endpoint. It is never transmitted to or stored on any other server.
MCP integration
The optional MCP server runs on your own machine and reads Markdown transcripts you explicitly export. It makes no network requests.
Your control
Delete any recording in the app and it is gone. Clearing your browser's site data removes everything Memo has ever stored. There are no backups because there is no cloud.
Questions about privacy? This is a Hatched Digital tool built with a simple principle: if we don’t have your data, we can’t lose it, sell it, or leak it.