Dictantor captures meetings and voice notes across Mac, iPhone, and Apple Watch. On Mac, it records microphone and system audio together and prompts you before meetings. Transcription runs on device in 25 European languages, separating your voice from everyone else. Search returns you to exact spoken moments. No Dictantor account or subscription. Supported recordings can sync through your private iCloud account.
I built Dictantor because I wanted a simple way to capture conversations and ideas without sending personal recordings to another company’s servers.
Dictantor started as an iPhone and Apple Watch voice notes app. Today, it’s also available on Mac, where it can record your microphone and system audio together, suggest recording before meetings, and turn conversations into searchable transcripts.
Everything is designed around a local-first approach:
• On-device transcription in 25 European languages • Your voice separated from everyone else • Transcript search linked to exact playback moments • No Dictantor account or subscription • Optional sync through your private iCloud account
The goal is simple: audio is the source, but useful, searchable text is the destination.
I’d love to hear how you currently capture meetings and voice notes, and what would make Dictantor more useful in your workflow. Thanks for checking it out!
I really dig the design and overall vision here. Awesome job @kremenskoy. I know the focus is on-device privacy but I'm curious if there's a way to allow users to allow their AI Assistants to search through the notes as well.
@gabe Thank you! I’ve actually been thinking about giving AI agents access to notes as well. A first step could be making transcripts easy to export and use with any LLM, and later I’d like to explore a more direct, opt-in way for agents to search through notes without losing the privacy-first approach.
And yes, Japanese is now on the list to investigate🕵🏻♂️ ✔️
Thanks for the suggestion!
Report
How well does the speaker separation work when several people are talking at the same time?
@lazar_j1 It works pretty well in normal conversations, but overlapping speech is still tricky. Right now the separation is fairly basic, it distinguishes between you and everyone else rather than identifying each speaker individually. Definitely something I want to improve 🛠️
I’ve been following this one for a while, and I’m really happy to see Dictantor finally launch! The local-first approach to storing and processing recordings is an awesome feature!
@paulmatvienko Thank you! Really appreciate you following along 💙 Keeping Dictantor local-first has been one of the main ideas from the start, so I’m really glad that part resonates.
You chose a no-account, no-subscription, iCloud-sync-optional model. What tradeoffs did that force (collaboration, cross-device reliability, support/updates), and how did you decide those were worth it versus a server-backed approach?
@curiouskitty I think the biggest tradeoff is that I have to keep the scope fairly focused. Right now, Dictantor is a personal tool within the Apple ecosystem, so there are no shared workspaces or collaboration features.
Sync depends on iCloud, which means I don’t control every part of it, and troubleshooting can sometimes be harder. Since the data stays local, I often have to rely on users describing what went wrong rather than being able to see it myself.
On the other hand, local transcription keeps the ongoing costs low and means recordings don’t need to go through my servers. That’s what makes the no-account, no-subscription approach possible.
For now, it feels like the right choice. I wanted Dictantor to be simple and private, and I’ll try to keep following that direction as it grows, even if some decisions change along the way.
Dictantor
Hey Product Hunt! 👋
I built Dictantor because I wanted a simple way to capture conversations and ideas without sending personal recordings to another company’s servers.
Dictantor started as an iPhone and Apple Watch voice notes app. Today, it’s also available on Mac, where it can record your microphone and system audio together, suggest recording before meetings, and turn conversations into searchable transcripts.
Everything is designed around a local-first approach:
• On-device transcription in 25 European languages
• Your voice separated from everyone else
• Transcript search linked to exact playback moments
• No Dictantor account or subscription
• Optional sync through your private iCloud account
The goal is simple: audio is the source, but useful, searchable text is the destination.
I’d love to hear how you currently capture meetings and voice notes, and what would make Dictantor more useful in your workflow. Thanks for checking it out!
Product Hunt
I really dig the design and overall vision here. Awesome job @kremenskoy. I know the focus is on-device privacy but I'm curious if there's a way to allow users to allow their AI Assistants to search through the notes as well.
Oh and please add Japanese :)
Congrats on the launch!
Dictantor
@gabe Thank you! I’ve actually been thinking about giving AI agents access to notes as well. A first step could be making transcripts easy to export and use with any LLM, and later I’d like to explore a more direct, opt-in way for agents to search through notes without losing the privacy-first approach.
And yes, Japanese is now on the list to investigate🕵🏻♂️ ✔️
Thanks for the suggestion!
How well does the speaker separation work when several people are talking at the same time?
Dictantor
@lazar_j1 It works pretty well in normal conversations, but overlapping speech is still tricky. Right now the separation is fairly basic, it distinguishes between you and everyone else rather than identifying each speaker individually. Definitely something I want to improve 🛠️
Loving how simple the UI is, good luck!
Dictantor
@artsmi Thanks! Really glad you like it 🙌
I’ve been following this one for a while, and I’m really happy to see Dictantor finally launch! The local-first approach to storing and processing recordings is an awesome feature!
Congrats on the launch, Alexander! 🚀
Dictantor
@paulmatvienko Thank you! Really appreciate you following along 💙 Keeping Dictantor local-first has been one of the main ideas from the start, so I’m really glad that part resonates.
Thanks for the support! 🙌
Mailwarm
Dictantor is a cool name. Points for that if I'm being honest..
Dictantor
@naimz Haha, thank you! Let’s just say the name was slightly inspired by the times we live in 😅
Product Hunt
Dictantor
@curiouskitty I think the biggest tradeoff is that I have to keep the scope fairly focused. Right now, Dictantor is a personal tool within the Apple ecosystem, so there are no shared workspaces or collaboration features.
Sync depends on iCloud, which means I don’t control every part of it, and troubleshooting can sometimes be harder. Since the data stays local, I often have to rely on users describing what went wrong rather than being able to see it myself.
On the other hand, local transcription keeps the ongoing costs low and means recordings don’t need to go through my servers. That’s what makes the no-account, no-subscription approach possible.
For now, it feels like the right choice. I wanted Dictantor to be simple and private, and I’ll try to keep following that direction as it grows, even if some decisions change along the way.