talat captures your microphone and system audio, transcribes both sides of every conversation in real time, and turns meetings into searchable, editable notes. It's powered entirely by your Mac's Neural Engine: your audio never leaves your machine. Choose custom LLM providers, write custom summarisation prompts, auto-export to Obsidian, push meeting data via webhooks, or query your history through an MCP server. It runs alongside Granola and other tools, so you can try it without switching.
Top comment
Hey Product Hunt! I'm Nick, and I built talat because I wanted Granola's magic without my audio living on someone else's servers. I've been obsessed with this space for about a year. It started when I discovered that macOS could tap system audio without recording video: something I'd never seen an app do before Granola. That led me down a rabbit hole into Apple's Core Audio taps API, and I ended up building an open source Swift library to make it more accessible. Over the past year I've been piecing together the puzzle: system audio capture, mic recording, acoustic echo cancellation, automatic meeting detection, custom notification windows. Recently discovering FluidAudio, which runs real-time transcription on the Apple Neural Engine, was the piece that brought it all together. It's early days and plenty of stuff needs work; speaker diarisation is rough, local LLM summaries can be hit and miss. Personally, the more I use talat, the less I care about perfect summaries and the more I care about the transcript just being there, ready to search and refer back to whenever I happen to need it. talat is a one-time purchase, and if you buy during pre-release you get app updates forever. I'd love your feedback: what works, what doesn't, what you'd want next. And if you already use Granola or another meeting tool, talat runs happily alongside it. You don't have to choose.
Comment highlights
This is the way. When the job can be done on local hardware you already own, it feels wasteful to rent offsite tokens.
I've used it for a few client calls and standups: transcription accuracy is solid (better than expected for local), custom LLM prompts let you tweak summaries exactly how you want, and the Obsidian export plus webhooks feel genuinely useful for power users.
This is a very thoughtful take on AI meeting notes. The fact that everything stays on the Mac makes the product immediately stand out, especially for users who care about privacy and control. I also like that you are positioning it as something that can work alongside existing tools instead of forcing people to switch. Do you see talat becoming more transcript-first over time, or is improving summary quality still a major focus?
really like the privacy-first direction here. having realtime meeting notes that stay fully on-device feels like a big win, especially for people who are not comfortable sending sensitive conversations to external servers. the obsidian export and custom prompts are a nice touch too. how has the response been so far from people already using granola or similar tools?
When someone is already using Granola (or a bot-based tool like Otter/Fireflies), what’s the exact “breaking point” that makes them switch to talat—and what does the migration look like in practice (history, exports, habits)?
The privacy angle here is underrated. Most notetakers treat "your data" like a byproduct. You're treating it like it belongs to you — because it does. What I'm curious about: do you think local-first transcription changes how people actually speak in meetings? Like, does knowing nothing leaves your machine shift the quality of what gets said?
I run meetings in two languages — some fully in Czech, some in English. Does the transcription handle both well, or is it optimized mainly for English?