Let’s make the first experience feel simple. You should be able to record a conversation and find your way back to any part of it.
Konos is in development Join the waitlist.
Local AI meeting notes.
Local transcription. No meeting bot.Product catch-up
A little more clarity
Today, 10:30 AM·3 speakers·03:24
Open the sample conversation
Exactly. I’ll share the updated onboarding screens by Thursday. There are a few details I want to simplify.
I can test the audio capture on Friday. I’ll check the microphone and system audio together, especially when someone joins late.
Great. I’ll put together the release checklist for Monday. Let’s keep all three tasks in one place so nothing gets lost.
One more thing: the speaker colors should carry through to the waveform. It makes a long conversation much easier to navigate.
That works. We can keep everything on the computer and let people choose a different model when they need one.
Perfect. That’s everything for today. I’ll catch up with you both next week.
A second look at your meeting.
Ask a question. Find the tasks. Keep the context.
Tasks from this conversation.
Room for your next steps.
Ask AI Chat for tasks, then add them here in one click.
00:12/ 03:24Highlight a voice
Record. Transcribe.
Chat. Follow up.
Both sides. One recording.
Your microphone and system audio, together. No meeting bot to invite.
Know who said what.
A readable transcript, speaker by speaker. One click takes you back to the moment.
Your meeting. Your questions.
Chat with a local model, Claude, or ChatGPT. Ask for the tasks you want to take forward.
Make room for what’s next.
Send the tasks from chat to your To-Do tab, all at once. Keep the original conversation close.
Find the speaker.
Follow the thread.
Choose a speaker and see their words light up across the waveform. Jump to the part you need, without replaying the whole conversation.
Try the speaker scrubberMeeting notes.
Your kind of AI.
Keep transcription and chat local, or choose an external AI provider.
Powerful models.
No cloud required.
Choose speech and language models that fit your hardware and language. Your recordings and transcripts stay in your own library.
Model performance depends on your hardware.
Claude
ChatGPT
Grok
Hermes
Pi
Cursor
Antigravity
OpenCodeA different model.
The same context.
Ask questions about your transcript with an external AI provider. Content you send is handled under that provider’s data policies.
How your data is handledKeep your meeting files together.
See what was on screen. Take your meeting files with you.
Screen context
See the moment behind a task.
Match a follow-up to what was on screen. Open the sample task to see its source.
Onboarding / First stepsA simpler start.
Task sourceYour workspace is ready.
Continue to workspace
Share the updated onboarding screens
“Exactly. I’ll share the updated onboarding screens by Thursday. There are a few details I want to simplify.”
View transcript at 00:38Meeting exports
Your files, in useful formats.
Keep everything in a Konos archive, or export just the part you need.
Konos archiveA little more clarity.kn
Audio, transcript and AI chat, together in one Konos file.
AudioA little more clarity.wav
Keep the original meeting audio as a WAV file.
TranscriptA little more clarity.md · .txt
Speaker-labelled text in Markdown or plain text, ready for your notes.
AI chatA little more clarity.md · .json
Keep your questions and AI responses as Markdown or JSON.
Your questions,
answered.
About pricing What is Konos?
Konos is a desktop AI meeting notetaker. Record microphone and system audio, read a speaker-labelled transcript, ask AI chat about the conversation, and collect the follow-ups in To-Do.
How does the website demo work?
The preview uses a prepared meeting, transcript, chat responses, and tasks. Local, Claude, and ChatGPT are illustrative model choices in the demo; switching between them does not call an AI provider or upload your files.
How do I turn the conversation into tasks?
In the sample workflow, AI chat presents the meeting follow-ups. Choose Add all to To-Do to collect them in the To-Do tab. Adding the same tasks again does not create duplicates.
What happens when I select a speaker?
Selecting a speaker highlights their transcript segments and fills the matching spans in the waveform. The speaker highlight is separate from the playback position, so you can see where that person spoke across the meeting.
Does meeting data leave my computer?
The desktop local workflow records, transcribes, and runs local-model chat on your computer. Choosing external AI processing can send selected meeting context to that provider, whose own policies apply. This website uses sample meeting content and does not accept meeting uploads.
Can I export my meeting files?
Export audio as WAV, transcripts as Markdown or plain text, and chat as Markdown or JSON. A Konos archive (.kn) bundles your audio, transcript, and chat. The website preview offers a sample Markdown export; available formats will be confirmed with each application release.
When can I download Konos, and what will it cost?
Konos downloads, platform details, and pricing are coming soon. Join the waitlist to register your interest and hear when everything is ready.
What happens when I join the waitlist?
Your email address is saved to register your interest in Konos. Joining does not create an account or commit you to a purchase. Confirmation and update emails are not currently sent automatically. The privacy page explains how signup data is stored.
Join the
Konos waitlist.
Everything is coming soon. Join the waitlist to hear when Konos is ready.