Hemory

About Hemory

Hemory (a name combining "hear" and "memory") is a continuous listening tool that turns everyday conversations into a private, searchable archive. It runs on an iPhone, Android phone, or Apple Watch you already own, auto-splits your day into moments with speaker labels, and stores the resulting text-not the original audio-in the cloud. You can connect it to AI tools like Claude, Codex, Cursor, or any agent supporting MCP to give those systems context from your past conversations.

Review

Hemory addresses a specific friction point: the gap between what gets said in a day and what actually gets captured and reused. Instead of relying on manual meeting recordings or scattered notes, it listens passively in the background and structures conversations into searchable text. The tool's MCP integration then lets external AI agents query that memory, which shifts it from a personal archive into something those agents can act on.

Key Features

• Continuous background listening (Max mode): One tap starts listening for up to 24 hours. The tool separates conversations by activity in real time, so you don't end up with one giant merged transcript.
• Auto-organized moments with speaker labels: Hemory splits your day into discrete conversations and applies speaker labels. Only detected speech counts toward your usage quota; silence doesn't consume it.
• MCP connectivity for AI agents: Connect Hemory to Claude, Codex, Cursor, OpenClaw, or other MCP-compatible tools. Your AI can then pull historical conversations as context-asking what was discussed yesterday, summarizing pricing talks from the past month, or drafting a half-year review from what it's heard.
• Multi-device support with no extra hardware: Runs on iPhone, Android, and Apple Watch. The makers report that on an Apple Watch Ultra 3 with other notifications off, 12 hours of listening leaves over 50% battery. Calls or other apps taking the microphone can interrupt recording; Hemory alerts you and tries to resume automatically, though a gap remains.
• Privacy-oriented audio handling: By default, original audio is not stored. It gets transcribed, then discarded. What persists in the cloud is the text memory, which the makers say has no current storage limit.

Pricing and Value

Pricing is not yet defined in the available reference material. The tool launched this week, and the makers have not published a pricing page or subscription tiers at this point. Usage quota is tied to detected speech, not total listening time, but specific quota limits or any free tier details are not stated.

Pros

• Captures spoken ideas and decisions without requiring manual note-taking or remembering to start a recording.
• MCP integration lets AI agents query your conversation history directly, which can feed into summaries, reports, or code generation.
• Runs on devices you already carry-iPhone, Android, or Apple Watch-so no dedicated hardware is needed.
• Speaker separation works in normal meeting rooms with 6-8 people, according to the founder's response to a user question.
• Deletion is straightforward: you can remove any captured memory, and deleting it also removes the associated audio from the recording device.

Cons

• Microphone conflicts with calls or video meetings interrupt recording on iPhone; the gap while the mic is unavailable means some conversation is missed unless you rely on a paired Apple Watch.
• Retrieval is based on keyword plus time, not semantic understanding-your AI does the connecting work, so the built-in search itself is fairly basic.
• Not well suited for people who only want occasional, manual recording of specific meetings rather than continuous background listening throughout the workday.

Hemory fits professionals who spend a large portion of their day in conversations and already work with AI agents that can consume context over MCP. It's less useful for someone who prefers deliberate, session-based recording or who doesn't use AI tooling that can query a memory store. The privacy model-transcribe and discard audio, keep text-will matter differently depending on your industry's consent requirements and your own comfort with continuous listening.

Open 'Hemory' Website