I'm the developer of Trace, a non-intrusive, shortcut-driven Mac app that records and transcribes your meetings on-device. I know, another meeting transcription app. Please bear with me though, I'm confident that this is at least a little novel.
I primarily built Trace for myself. I'd been using MacWhisper, but there was enough fiddling before each call that I'd forget to start it and walk out of an hour-long meeting with nothing written down. So the things I cared about most were that it's quick to activate and stays out of the way. You activate Trace by pressing a global shortcut (configurable), which reveals a small bar at the bottom of your screen (there's also a keystroke and/or option to hide it entirely if you'd rather not see it at all).
As I was building it I wanted to bake in a couple of workflows I'd wished for in other transcription apps.
1. Mid-meeting you can press another global shortcut to mark a "key moment" and type a note. The note shows up in the resulting transcript inline at that timestamp. I wanted to add this because I kept catching myself thinking "wait, that bit matters" in meetings and reaching to jot it down in a separate app like Obsidian, which I then needed to add context to, which took me out of the meeting. I use it all the time. If I paste the transcript into an LLM afterwards (which I find myself doing more and more these days) the important moments are flagged so it doesn't gloss over them. This is more noticeable in longer meetings with lots of topics. 2. With another keyboard shortcut you can summon a rough live recap (subtitles, basically) to quickly recap what's just been said.
Trace uses standard macOS microphone and system recording APIs to capture both sides of the conversation as two separate tracks and then runs the system side through on-device diarization to identify speakers. Right now we only label them as "Speaker 1", "Speaker 2", etc but there are plans for speaker labelling in the future. You can also show a "live recap" as the call is happening to review what someone just said.
All transcription models run on your machine. To be clear though, Trace doesn't do any of the summarising itself, it just produces a markdown transcript, so if you want summaries then you need to pass the output to an AI.
The app is sandboxed and your audio/transcripts are never uploaded anywhere - they just exist as audio files and markdown on disk. The only network call Trace is required to make is on the first run to download the speech and speaker models (around 500MB) from Hugging Face, and after that it can be used fully offline. If enabled, a Google Calendar integration can auto-name sessions but that needs a network connection.
The app is Ā£9.99 on the macOS App Store. I've been using it every day for months now and I'm super happy with how it's improved my workflow. Feedback very welcome.
Nice! I really like how many variations on this idea are coming out. MacWhisper used to be great, but is kinda of a buggy mess now.
I'm making my own, for personal use. I did a survey of many and they all (that I could find) skip the fundamentals.
The major issues that I've run into:
- Crash recovery. Most of these apps are incredibly buggy and crash all the time, taking the recorded audio with them. Macwhisper is incredibly bad at this.
- Disk space. Many of these apps save wav files to disk. After a few hours of meetings, you may end up with gigabytes eaten.
- Microphone bleed. People don't always use headphones, the system mic will pick up the speaker sounds, causing duplicate (approximately) transcriptions.
I've yet to find a solution that handles all these correctly, let alone having high quality transcriptions.
Anyway, most of these apps are built around https://github.com/FluidInference/FluidAudio, if anyone is curious. Their readme has a big list of similar apps as well.
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.
For crash resilient data, you have a few options:
- Journaling file structures (telegraph what you're about to write, then write it, then signal completion)
- memmap your important data structures to a file (they will be flushed to disk no matter how your app dies - short of a power loss)
- post-crash dump (put last-minute writers in a crash handler to save it to disk)
A journaling file structure is the most secure, because it's designed with the assumption that writing will eventually fail. memmapped structs are easy and cheap, and get you 99% of the way there (only power loss will lose your data). Crash-time writing is doable with a crash handler like KSCrash, but there are many ways an app can crash without triggering a crash handler (thermal kill, exceeding quota, memory jetsam, etc). You also need to write your data in a signal-safe manner.
I had the same experience so started building my own. All problems are solvable, just working on the polish.
- crash recovery: part one is use ADTS aac (even if process crashes, audio is saved up until it does). Part two is isolating the transcription/summaries in separate XPC services.
- disk space: AAC 64kbps mono soles it. Could use Opus for further reduction but both are small.
- speaker bleed: macOS voice isolation processing solves this. Itās a nightmare to get setup, but works great once done.
- library: using argmax SDK - by a bunch of ex-Apple on device AI folks.
It it wasnāt for CoreAudio, Iād say it was easy to make. Argmax, Whisper, and llama.cpp - wrapped in the right architecture, mostly just work.
Iām having fun nerding out on the details like custom vocabulary (get the names of the people in here meeting right), inferring speaker names from transcript, calendar integration, nice UI, etc.
Nice tip on FluidAudio that's the kind of thing I've been looking for. Thanks!
Iām using MacParakeet these days. If your language is supported, definitely give it a try. Itās much faster and lower footprint
The App looks really interesting and Iād love to try it out. How well does it work in other languages than English? For me, German would be important.
Due to audio quality, transcription sometimes produces garbled output or understands something wrong. FluidVoice offers the option to use a LLM to āinterpretā the text to rescue garbled audio through context. Do you also plan to support something like this? This would be a great feature!
Would love to use this app, I recently thought about coding something similar myself. I would need to only record my own voice due to privacy laws (here in Germany, you can record yourself without consent). With overthe-ear headset, the microphone only captures my voice. Would need to store the original audio plus the transcription. Ideally, you can configure it to start as soon as it detects a new window with a given title (i.e. Webex launches meetings in a new window named "Meeting ....").
Had the same idea, but have to focus on my main business. This comes at the right time!
I just purchased it. What's the best way to give you feedback? (Do you want any?)
From the top of my head: - will the mic switch automatically when I am at my office? Or do I have to change settings every time? Maybe a preference of what's available + auto switch would be good. - I personally don't need the hot key. Menu bar icon would be fine. - Download the model is a long process. Put it into the installer, not into the bar on the bottom - Speaker correction would be amazing. If it could "Learn" the speakers based on voice. - Overall neat app. Good animations and UX
**Speaker 1** [00:00] What if I fell to the floor?
**Microphone** [00:02] Yes, this is Phil, I'm just speaking, this should be my voice, and there's music in the
**Speaker 1** [00:05] Couldn't tell this anymoreThanks for the feedback. Feel free to drop an email to hello@traceapp.info 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.
Learning speakers is also on the to-do list.
P.S. Great choice in test audio. What a banger.
I'd love to have a purchase option not tied to the App Store if possible. I don't use an Apple account with my Mac, but I would love to try Trace.
This is definitely on the to-do list if thereās 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 Iāll bump this up the priority list.
Agreed, no need to tie it into Apple either.
Also agreed, my work prohibits App Store apps so i have to skip things like this.
The key moments feat is neat. Been working on a free opensource offline transcriber that runs fast on CPU and does diarization too
This is an excellent product and exactly what I've been looking for. But most of my meetings are done on my company Mac, and they definitely won't let me install this kind of software, even though I'd be willing to pay for it myself.
And if it runs on the browser without install it would not probably be able to record your other browser (or app) audio
This looks sick. I was going to download it but for $10 I am more willing to attempt asking Claude to implement something like it, than to purchase.
I would be more willing to purchase if it was open source and I could build from source to try it first.
It's kinda funny how frontier LLMs change the game when it comes to software. If it becomes so good to make whatever little utility you want, why would I pay 10 dollars when an AI subscription is 20 bucks and I can build way more in a month for that $20? Especially since it's very likely people on show HN have simply used AI anyway, so why would I pay for your prompts?
I don't really recommend it. If the software is a one-time purchase, there's no need to rewrite it with an LLM. Rewriting the tokens could cost more than just $10.
* full price tokens, yes
Not the subsidized subs
I'd much rather spend $10 than have to sit at a prompt every day babysitting the thing, after working all day sitting at a prompt babysitting other things
Get a daily email with the the top stories from Hacker News. No spam, unsubscribe at any time.