cd /news/artificial-intelligence/ai-meeting-notes-when-you-re-hard-of… · home topics artificial-intelligence article
[ARTICLE · art-60737] src=zackproser.com ↗ pub= topic=artificial-intelligence verified=true sentiment=↑ positive

AI Meeting Notes When You're Hard of Hearing: My Working Setup

Zack Proser, a WorkOS engineer who is partly hard of hearing, uses Granola in every work meeting to capture live transcripts and AI-generated notes, reducing the cognitive load of capturing meetings to a single button. The workflow allows him to quietly recheck missed phrases mid-meeting via the live transcript and review decisions, owners, and details afterward, avoiding the social cost of asking for repeats. Proser also uses Granola as a connector for AI systems he builds at WorkOS, enabling call-to-agent workflows without manual copy-paste.

read6 min views15 publishedJul 13, 2026
AI Meeting Notes When You're Hard of Hearing: My Working Setup
Image: Zackproser (auto-discovered)

Three questions. One honest recommendation.

Not sure which tool fits? Ask the advisor — a question or two, then a straight answer.AI Meeting Notes When You're Hard of Hearing: My Working Setup

I'm partly hard of hearing, and meetings are where that costs me the most. Crosstalk, quiet speakers, bad laptop microphones, and compressed call audio can turn one sentence into a guessing exercise. This happens even when I'm paying full attention.

I use Granola in every work meeting and consulting call. The workflow reduces the cognitive load of capturing a meeting to one button: I start the meeting in Granola, then I listen. If I miss a phrase, I can quietly recheck the live transcript while the conversation continues. After the call, I have the decisions, details, and owners in one place.

That mid-meeting recheck is the part that matters most to me. It catches the miss without making me stop the room and ask someone to repeat a sentence I may have understood correctly.

How I use a live transcript to recheck what I misheard

When I lose a sentence, I usually still have enough context to know that something important was said. Maybe a date changed. Maybe someone assigned an action item. Maybe a quiet speaker gave the one constraint that changes the whole project.

My old options were to interrupt, guess, or hope the detail came up again. Now I use this short loop:

  • I notice that I missed a word, number, name, or decision.
  • I open the current transcript in Granola.
  • I check the last exchange while the meeting keeps moving.
  • If the transcript resolves it, I return to the conversation. If it does not, I ask.

This is useful even on a good hearing day. Full attention cannot fix overlapping voices or a microphone that clips the beginning of a sentence. The transcript gives me a second path to the information.

I also avoid the social calculation that comes with every request to repeat something: Was that important enough to stop everyone? Did I actually mishear it? Has someone already moved to the next point? Quietly checking the transcript removes that calculation when the text is clear.

Get 3 Months Free How AI meeting notes help me catch what I missed afterward

The live recheck handles the immediate gap. The captured record handles everything I do not realize I missed.

After a call, I review the notes for three things:

  • What did we agree to?
  • Who owns each follow-up?
  • Which names, dates, requirements, or constraints will matter later?

I no longer have to reconstruct those details from memory. That matters because a plausible memory can still be wrong. A clean record lets me check the source before I send a follow-up or start the work.

I use the same record as input to the AI systems I build at WorkOS. Granola is one of the first connectors I wire into a new agentic system. I can finish a call and ask a blog bot to draft several posts from that meeting. The path is call to notes to agent, with no copy-paste step in the middle.

For a closer look at the product and the rest of my daily workflow, read my Granola review. Sharing meeting notes without rebuilding the conversation from memory

Sometimes another person needs the context: a teammate who missed the call, a client confirming the plan, or a colleague taking over an action item. Granola gives me a one-click way to share the notes or transcript.

The practical benefit is accuracy. I can share what was captured instead of writing a summary from memory and wondering whether I dropped the detail that the other person needed. Everyone can refer to the same decisions and owners.

Accurate records matter in professions where a missed phrase carries extra weight. The workflows differ, but court reporters handling intake and scope calls and doctors trying to stay present in conversations face a related problem: listening and documenting compete for attention.

My meeting notes app setup as a hard-of-hearing professional

My setup is intentionally small:

  • Before the conversation starts, I open Granola and start the meeting.
  • During the meeting, I listen and contribute instead of trying to type a parallel record.
  • When I mishear something, I check the live transcript.
  • After the meeting, I review decisions and ownership while the conversation is fresh.
  • I share the notes or pass them into the system that needs them next.

The habit matters more than elaborate configuration. If I forget to start capture, the safety net is gone. Opening Granola at the start of every meeting makes the rest of the workflow dependable.

Where this setup stops helping

Granola is a working aid in my setup. I do not treat it as a formal accommodation, an accessibility certification, or a replacement for the support another deaf or hard-of-hearing person needs.

Transcripts can also be wrong. Crosstalk, quiet talkers, names, and poor audio can cause trouble for software too. If exact wording carries legal, medical, financial, or personal consequences, I verify it. If the live transcript does not settle what I heard, I ask the speaker to repeat it.

My experience is one person's experience. Some people need live captions, CART, an interpreter, an assistive listening system, or another accommodation. My setup works for me because it adds a quick second reference during the meeting and a reviewable record afterward.

FAQ

Can I use an AI transcript when I mishear something in a meeting?

Yes. In my workflow, I check Granola's current transcript when I miss a phrase, number, or name. It lets me recheck the exchange without interrupting. I still ask for clarification when the transcript is uncertain.

Does this help if I was paying full attention?

Yes. Hearing loss is not an attention problem. Crosstalk, low voices, room acoustics, and bad call audio can hide words even when I am focused on the speaker. The transcript gives me another reference.

Is Granola a replacement for captions, CART, or an interpreter?

I would not use it that way. I use Granola for meeting capture, mid-meeting rechecks, and review afterward. The right accessibility support depends on the person and the setting.

What should I review after the meeting?

I check decisions, action items, owners, dates, and any detail I was unsure about during the call. I do this while the conversation is still fresh, then share the record with anyone who needs it.

If you want to try the same setup, new users get their first three months free.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @granola 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/ai-meeting-notes-whe…] indexed:0 read:6min 2026-07-13 ·