It wouldn't be too difficult, but I'd be hesitant to do it for fear of bloating what Trace was designed to do. Trace is built to do transcription and hand you clean markdown (which you can paste into whatever LLM you prefer for a summary). Many alternative solutions in this space are bolting on lots of "AI frills", which, in my experience, is what makes them start to feel heavy.
Looks like echo cancellation is to blame here. I'm working on a more permanent fix but for now try turning off "Cancel system audio from the microphone" to see if that helps.
Thanks. Trace integrates with your Mac calendar, so if you’ve signed into your accounts with those it will pick them up. No support currently for direct Teams/Exchange integration but I’ll add it to the list.
Thanks for the support. The videos are Vue components with CSS animations. As mentioned in a previous reply I’m happy to share the component if it’s of interest.
Recording and consent rules vary quite a bit by country and region, so it's on whoever is doing the recording to get the consent they need and to follow the law where they are. Still, I like the idea of putting a note on the FAQ to make that clear.
Thanks for the feedback. Feel free to drop an email to [email protected] if you want a chat. Happy to hop on a call too if you'd prefer.
For the switching, do you mean if you hot-swap during a call? The mic should auto-switch if you've got System default selected, but feel free to give it a go and report back. If it doesn't do what we expect I can absolutely take a look at changing the behaviour.
Trace has two engines that you can choose from. The fast one uses NVIDIA's Parakeet-TDT 0.6b v3 model run through FluidAudio, which surprised me with how fast it was. There is also an accurate engine, which uses Whisper large-v3-turbo via WhisperKit, which is slower but holds up better on accents and jargon.
Crash recovery is definitely something that I want to spend a bit more time on. I'm not entirely sure how Trace handles crashing right in the middle of a recording, so I'm going to put a bit of time aside in the next few days to properly explore this and see if I can come up with an elegant solution to it.
I think I've got the other two bits covered. I pushed an update yesterday that adds active echo cancellation so that audio playing through the speakers (or leaky headphones) won't get transcribed twice if it is picked up by the microphone. It can be disabled in preferences, but it's on by default.
The disk space issue is one that I considered as well. By default, Trace deletes the actual audio recordings as soon as transcription is successfully completed, so the idea is you keep just the markdown transcript rather than the gigabytes of raw audio. If you want, there's a preference to disable the auto-deletion. There's a bit more on the support page here https://traceapp.info/support (search for "Auto-deletion of audio").
FluidAudio is a big part of this and is actually used in two places during a session. It runs the Parakeet EOU model for the instant recap (which isn't hugely accurate, but it's good enough for the job) and after the call it's also used to transcribe the recording, depending on which engine you've selected (Trace offers a fast and an accurate one). If the fast engine is selected, we use FluidAudio with the Parakeet-TDT 0.6b v3 model for transcription, which then goes through Pyannote and WeSpeaker for diarization. If the accurate engine is selected, we use WhisperKit with the Whisper large-v3-turbo model for transcription, and SpeakerKit for diarization.
This is definitely on the to-do list if there’s enough demand for it. The payment/distribution/updates infra required is not insignificant, especially if nobody was that bothered, but by the sounds of it they are so I’ll bump this up the priority list.
It does indeed. Trace will record your system audio regardless of your speaker volume. You do have the option to mute your own mic temporarily though, via a button on the “pill” or a global keyboard shortcut.
I've been working on a minimal, keyboard-driven Mac app that records and transcribes your meetings on device, similar to MacWhisper but way more minimal and simple. I originally built it for myself because I kept forgetting to start MacWhisper before calls. It uses Apple APIs to capture the mic and system audio, diarises it, and gives you a markdown transcript at the end. What I quite like about it is that during the call you can mark key moments, which then get inserted into the transcript at the right time. So if you do hand it to an LLM to summarise, it knows what was important.
I like the idea, but this does raise a larger question about sensitive information (that you've alluded to, but I think it's more of a systemic/cultural problem). Everyone knows you shouldn't be pasting env secrets and customer data into Claude, but many do it anyway because the perceived productivity gains offered by tools like Claude Code are just too high to ignore (and the payoff is instant and the potential risk is later).
What this does though is make that visible. If someone partakes in that bad practice, then the secrets are already sitting in the transcripts on their laptop - this just moves them somewhere shared and permanent. Maybe that's enough to discourage the practice, idk. Still, not strictly your problem, but an interesting one nonetheless.
I'm getting a 404 on the link. I've always struggled with how to track open "threads" with colleagues, so I hope this does what I think it does. Interested to see how it's implemented.