Smart Context · 7 min read

Dictating Into Mail vs. Notes vs. Slack: A Practical Test

I dictated the same paragraph into five apps. The output was different every time.

This was a test. I wanted to see how Smart Context actually performs across different apps. So I dictated the same rough paragraph into Mail, Slack, Notes, Pages, and a code editor.

The results were surprisingly good , and surprisingly different.

The test

Here's what I dictated (rough, casual, my normal voice):

"so basically the thing is we need to move the launch date back by like two weeks because the QA team found some issues with the payment flow and we can't really ship that so yeah thursday is out"

Here's what came out in each app:

In Mail: We need to move the launch date back by two weeks. The QA team found issues with the payment flow that we can't ship. Thursday is no longer viable.

In Slack: launch moving back ~2 weeks , QA found payment flow issues, can't ship. thursday's out

In Notes: Launch date moving back two weeks. QA found payment flow issues. Thursday is out.

In Pages: The launch date will be postponed by approximately two weeks due to payment flow issues identified by the QA team. Thursday is no longer a viable launch date.

In code editor: so basically the thing is we need to move the launch date back by like two weeks because the QA team found some issues with the payment flow and we can't really ship that so yeah thursday is out

Same input. Five different outputs. All automatic.

What I learned

Mail gave me a professional email. Complete sentences, proper structure, no casual language. Ready to send.

Slack gave me a casual message. Short, direct, no sign-off. Exactly how I'd write in Slack manually.

Notes gave me structured bullets. Clean, readable, good for reference.

Pages gave me formal prose. Appropriate for a document or report.

Code editor gave me exactly what I said. No polish. Because code comments should sound like the developer, not like a marketing team.

Apps tested , five different outputs from the same input, all appropriate for the destination

The one that surprised me

The code editor result surprised me. I expected some polish. Instead, Smart Context detected I was in a code editor and left my words almost untouched.

That's the right call. Code comments should be precise and technical. Polishing them into fluent prose would destroy their value.

I tested this with a technical comment: "TODO: refactor this to use the new API endpoint /v2/users instead of /v1." It preserved exactly. No "improvement." No "enhancement." Just my words.

Where it gets it wrong

Smart Context isn't perfect. Here's where I override it:

Casual emails to close collaborators: The Mail profile is sometimes too formal. I override to "friendly" for people I know well.
Formal Slack channels: Some Slack channels are more professional than others. I override to "professional" for client-facing channels.
Mixed-use apps: Notion can be notes or docs. I set it to "notes" by default and override when needed.

I override maybe 1 in 10 dictations. The other 9 are right on.

The honest note

Smart Context is a default, not a dictator. If the automatic tone isn't right, override it. The goal is to reduce editing, not eliminate your judgment.

The bottom line

The same paragraph should not come out the same way in every app. Smart Context makes that automatic.

Dictate naturally. Let the app handle the tone. Override when needed. That's the workflow.

Try Echo Flow free for 14 days · See how Smart Context works · Why context-aware dictation

Sources

Echo Flow's SmartContextService: detects the frontmost app and resolves a per-app tone profile.
Writer's own testing: across Mail, Slack, Notes, Pages, and code editors.
Apple AXUIElement / accessibility APIs: used to observe the active application locally.