Now on iOS 17 · 4.8 · 12,847 ratingsv3.4
v3.4 · on-device transcription in 12 languages

Voice memos
with structure.

Echolab records, transcribes, and routes your spoken thoughts into the tools you already use. 412k thinkers ship more thanks to a microphone.

4.8
12,847 ratings
0
Memos / quarter
14:42 REC · 0:47
Memo · Tuesday brain dump

"Move the auth flow…"

0:4702:18
14:42:08 Move the auth flow into its own service. Tell @maja we'll review Thursday — → Linear ENG-412
RECORD
LIBRARY
FLOW
Sent → Linear in 2.4s
The VergeWiredStratecheryDaring FireballFoundry 47Mercury OSField Notes Co.Plover StudioNorthwind LabsOn DeckThe VergeWiredStratecheryDaring FireballFoundry 47Mercury OSField Notes Co.Plover StudioNorthwind LabsOn Deck
iOS · Android · macOS · Web · 47 languages

A microphone
that finishes
the thought.

Four screens. The audio gets recorded, transcribed, structured, and delivered.

01 · Record

Tap and talk

One tap, no signup, no warm-up. 1.4M memos recorded last quarter, average length 47 seconds.

Avg memo · 47s
02 · Transcript

Smart transcripts

94.6% accuracy in 47 languages. Speaker diarization included. Edits sync back to audio.

[Hiroshi] So if we ship the auth split this week —
[Elena] We probably need to bump the migration to Friday.
[Hiroshi] Agreed. Tag @maja and @rafael.
94.6% confident · 0:47–01:12
03 · Workflows

Workflows

Send memos to Notion, Linear, Things, Gmail. The 'Tuesday brain dump' rule built in 12 seconds.

Notion
Linear
Things
Gmail
Slack
Roam
Obsidian
Webhook
04 · Vault

Vault

Encrypted at rest, deletable per memo. We don't train models on your voice. Ever.

encryption: AES-256
training_use: 0.0%
retention: user-controlled
vault_pin: optional per-memo
Three voices · three workflows

Talk fast.
Ship faster.

An app for people who think out loud. Recorded, structured, routed, archived.

01 / 03Workflow

Rafael Romero

Solo founder, Sevilla

Records every walk. The 'product memo' workflow lands as a 47-line Linear backlog by the time he's home.

02 / 03Workflow

Adelaida Pérez-Costa

Therapist, Madrid

Voice notes after each session. Encrypted vault means HIPAA-grade peace of mind without HIPAA-grade cost.

03 / 03Workflow

Maja Nilsen

Editor, Northwind Press

Four-person interview transcribed with speaker diarization in 2.4 minutes. Used to take 4 hours.

Not Voice Memos.
Not OtterAI.

A tool for people who think faster than they type and don't want to bargain with privacy.

01
Voice memos that sit forever
Memos that route to your tools
02
Walls of unstructured text
Headed, bulleted, summarized
03
Generic transcript accuracy
94.6% with speaker tags
04
No language support
47 languages, switch mid-memo
05
Sold to AI training pools
Encrypted, never trained on
User reviews · App Store

"Echolab replaced my notes app. I record while walking, the structured note lands in Notion before I'm home."

Hiroshi Tanaka
Founder, Plover Studio
4.8
App Store rating
12k+
Downloads / week
47s
Avg memo length
94.6%
Transcript accuracy

Pricing
without trickery.

No per-minute math. Cancel in the App Store.

Free

$0
30 minutes / month
  • 30 transcribed minutes
  • 10 memo library
  • Basic workflows
  • On-device storage

Pro

Most picked
$5/mo
Per month · most picked
  • Unlimited memos
  • Speaker diarization
  • Custom workflows
  • Encrypted vault
  • Notion / Linear sync

Studio

$9/mo
For teams of 5
  • Everything in Pro
  • Shared library
  • Team workflows
  • Audit log
  • SAML SSO

Asked
often.

How accurate is the transcription?

94.6% on clean audio across 47 languages. Drops to ~84% on noisy outdoor recordings; we add a 'low confidence' marker on each affected segment so you know what to verify.

Is my voice used to train models?

No. We've taken a public pledge: zero customer audio is ever used to train Echolab models. Your audio is encrypted at rest and deleted on request.

Does it work offline?

Recording works offline. Transcription runs on-device in 12 languages on iPhone 14+ and Pixel 7+. The other 35 languages need a brief cloud round-trip.

Can it integrate with my tools?

Yes. Notion, Linear, Things, Asana, Gmail, Slack, Roam, Obsidian, plus a generic webhook. Pro users build custom workflows in 12 seconds.

What's new

Version 3.4

Releases biweekly. Driven by power users in the Echolab Slack.

v3.4Apr 22

On-device transcription extends to 12 languages including Japanese and Portuguese.

v3.3Apr 02

Custom workflows: build your own routing rules.

v3.2Mar 11

Speaker diarization moves to v2 model. Up to 8 voices reliably tagged.

v3.1Feb 18

Encrypted vault. Per-memo PIN for sensitive recordings.

Download free.
Talk to ship.

30 free minutes a month. No card required.