Echoic

Remember every meeting without inviting a bot.

Record meetings, get private transcripts, and dictate into any Mac app. Speech processing stays on your device by default.

No bot joins your callsAudio transcribes locally by defaultNo account required to start

Free for macOS 14.2+ · Apple Silicon recommended

Built for meetings in classrooms, labs, clinics, offices, and teams.

Classrooms
Labs
Clinics
Client work
Teams

Start recording in seconds.

Download Echoic, grant microphone permissions, and capture the next conversation without inviting a bot.

01

Download & Install

Download the DMG, drag Echoic to Applications. It lives in your menu bar.

02

Set Your Hotkey

Pick any key combo. That's your dictation trigger — works in every app on your Mac.

03

Hold & Speak

Hold your hotkey, talk, release. Text appears at your cursor instantly.

What you'll need

  • A Mac running macOS 14.2 or later
  • Apple Silicon recommended
  • ~600 MB free storage for speech models
  • Microphone and Accessibility permissions

Capture the conversation. Keep the context.

Echoic starts with meeting recall, then adds dictation and AI assistance for the moments when you need to turn speech into work.

How it works

Start recording, stay present, leave with notes.

1

Record

Microphone and system audio are captured together.

2

Transcript

Speaker-labeled speech appears while you stay in the conversation.

3

Notes

Summaries, exports, and action items are ready after the meeting.

Meeting mode

Transcripts, summaries, action items, and exports.

  • Dual-Stream CaptureRecords microphone and system audio simultaneously — your voice and everyone else, cleanly separated.
  • Live TranscriptWatch the conversation unfold in real time with speaker-labeled chat bubbles.
  • Speaker DiarizationIdentifies and separates speakers into "Me" and "Them" channels with inline renaming.
  • AI Summaries & Action ItemsGet meeting summaries and extracted action items from any AI provider you choose.
  • Calendar IntegrationSee upcoming meetings in the menu bar. Quick-start recording from any calendar event.
  • Multi-Format ExportExport as Markdown, DOCX, or PDF. Markdown is Obsidian-vault compatible.

Dictation mode

Speak into any app on your Mac.

  • Type Into Any AppHold your hotkey, speak, and text appears at your cursor — in VS Code, Slack, Terminal, anywhere.
  • Sub-500ms LatencyNear-instant transcription with local neural models. No round-trip to the cloud.
  • Voice Commands"New line", "delete that" — natural editing commands built in.
  • Concurrent ModeDictate while recording a meeting — both modes run simultaneously without interference.

Private by default, plain about how it works.

Echoic keeps the privacy story concrete: speech processing happens on your Mac by default, meetings do not need bots, and audio is not uploaded for transcription unless you choose a workflow that sends data elsewhere.

On-device speech

Speech recognition runs on your Mac by default.

No meeting bot

Capture system audio without adding another attendee to the call.

No audio upload by default

Meeting audio does not need to leave your machine for transcription.

No Audio Uploaded

Speech recognition runs on your device by default, so meeting audio is not uploaded for transcription.

Designed for sensitive workflows

Local-first processing can help reduce exposure of sensitive audio. Echoic is not a compliance certification by itself.

What leaves your Mac?

Data typeDefaultIf you connect cloud AI
AudioStays on your Mac for local transcription.Not sent for transcription by default.
TranscriptStored locally in Echoic workflows.May be sent to the AI provider you choose for summaries, cleanup, or chat.
Meeting notesGenerated and stored locally when using local AI.May be generated by the selected provider from transcript text.
Calendar metadataUses local or Apple calendar permissions you grant.Depends on the calendar account you connect in macOS.

Simple Pricing

Start Free. Go Pro When Ready.

Start free for light use. Upgrade when regular recording, advanced dictation, and custom AI workflows become part of your week.

Local speech engines10+ AI providersUnlimited dictationEvery export format

Free

$0forever free

Best for trying Echoic or light personal use

10 recordings or raw audio transcriptions per month

  • Import existing notes and transcripts anytime
  • Unlimited local history
  • Cross-meeting search and chat
  • System-wide plain dictation
  • Core local speech engines
  • AI summaries & action items
  • Apple Calendar
  • iCloud syncsoon
  • All export formats (MD, DOCX, PDF)

Everything runs locally by default. No account required.

Download Free for macOS
Most Popular

Pro

$79/year

Less than $7/mo

Best for professionals who record regularly

Upgrade when you need unlimited recordings, advanced dictation, custom AI workflows, and priority support.

  • Everything in Free
  • Unlimited recordings and raw audio imports
  • Advanced local speech engines
  • Custom AI templates
  • Advanced dictation commands
  • Code command dictation
  • Priority support

Coming soon, included in Pro

  • Obsidian sync
  • Google & Outlook Calendar sync
  • Speaker research dossiers
  • CLI bridge for agents, Hermes & OpenClaw

Annual subscription, billed yearly. Renews automatically. Cancel anytime.

Go Pro — $79/year
No credit card required to start
Cancel anytime, no questions
Mac App Store coming soon
BYO AI Provider

Local first. Bring your provider when you need more.

Get summaries, action items, and answers from transcripts. Pick the routing mode first; provider details stay available for power users.

Fully local

Use Ollama or Apple Intelligence when transcripts should stay on your Mac.

Bring your own API key

Connect OpenAI, Anthropic, Gemini, OpenRouter, Groq, xAI, or DeepSeek.

Use external login

Route through Codex Login when you manage that account in Terminal.

Ollama logo
Local

Ollama

Fully local, 100% offline AI

OpenAI logo
Cloud

OpenAI

GPT-4o and beyond

Codex Login logo
External login

Codex Login

Use your local Codex CLI login

Anthropic logo
Cloud

Anthropic

Claude 3.5 Sonnet

Groq logo
Cloud

Groq

Ultra-fast inference

xAI logo
Cloud

xAI

Grok models

Apple Intelligence logo
Local

Apple Intelligence

On-device Apple AI

OpenRouter logo
Cloud

OpenRouter

Any model, one API

Gemini logo
Cloud

Gemini

Google Gemini models

DeepSeek logo
Cloud

DeepSeek

DeepSeek R1 & V3

What AI can do in Echoic:

Meeting SummariesAction Item ExtractionCustom AI TemplatesDictation CleanupPer-Feature RoutingStreaming Generation

Local AI mode available

No internet connection required

Use Ollama locally — zero data leaves your Mac

Choose speed, accuracy, language support, or zero-download Apple Speech.

Every speech engine runs locally on your Mac. The technical model names are available when you want to tune the tradeoff.

Fastest EnglishParakeet v3
Most languagesWhisper
Smallest downloadMoonshine
No model downloadApple Speech

FAQ

Questions people ask before recording.

Clear answers about privacy, meeting capture, dictation, and Mac compatibility.

What is Echoic?

Echoic is a private voice intelligence app for Mac. It records meetings, creates transcripts and notes, and lets you dictate into Mac apps.

Does Echoic work without a meeting bot?

Yes. Echoic runs as a native Mac app, so you can capture conversations without inviting a visible meeting bot into the call.

Is Echoic private?

Echoic is designed around local-first speech processing and user-controlled AI providers, so you can keep transcription on your Mac by default.

Does Echoic upload my meeting audio?

Speech recognition runs on your Mac by default, so meeting audio does not need to be uploaded for transcription. Optional AI features may use providers you choose.

When does my data leave my Mac?

Audio stays on your Mac for local transcription by default. Transcript text may leave your Mac only when you route summaries, cleanup, or chat to a cloud AI provider you choose.

Does Echoic work with Zoom, Google Meet, Teams, and FaceTime?

Yes. Echoic records from your Mac instead of joining as a meeting bot, so it is designed to work with common meeting apps and browser calls that play through your system audio.