Troubleshooting and FAQ

Microphone access is denied

If the menu shows an orange line “Microphone access is off – open Privacy settings…”, Minutes Ago has no microphone permission. Choose that line to open System Settings, and turn on Minutes Ago in Privacy & Security › Microphone.

If the first line of the menu reads “Error: no input device”, no microphone was found. Connect one and choose it in Settings › Recording › Microphone (or with the microphone chip in the Recordings & Minutes window).

The microphone could not start

If the menu shows the orange line “The microphone could not start (…) · Try Again”, with “Trying again in n s” below it, the microphone failed to start (this could happen right after the Mac woke from sleep). Minutes Ago tries again by itself (after 15 seconds, 30 seconds and 1 minute, then up to every 5 minutes), and at once when an input device is connected or changed. Choose Try Again to start it now; holding the pointer over it shows the last error.

Meanwhile the first line of the menu reads “Not recording”, the meeting bar shows the same orange line, and the menu bar icon shows a triangle.
The app audio of the app you chose in App Audio is still recorded.
Minutes Ago stops recording before the Mac sleeps and starts again after it wakes.

The microphone changed to a Bluetooth device

With System Default chosen as the microphone (Settings › Recording, the microphone chip, or Start Meeting…), Minutes Ago follows the Mac’s default input, for example when you plug in headphones. When the default input becomes a Bluetooth device, Minutes Ago follows it, and the menu’s warning starts with “System default input is now name.” with a button “Use name” for another microphone (built-in, then USB). Choosing it sets the microphone to that device, so it no longer follows the default. Bluetooth microphones often deliver narrow-band audio and can drop out; a built-in, USB or wired microphone gives better recordings and transcripts.

A microphone you chose that is not connected now is shown as “name (Not Connected)”.

Nothing is recorded: “Microphone and App Audio are both None”

If the menu reads “Not Recording · Microphone and App Audio are both None”, or shows the orange line “Microphone and App Audio are both None · Nothing is recorded”, both sources are set to None. Choose Use the Microphone in that line, or choose a microphone in Settings › Recording › Microphone or an app in App Audio ▸.

App audio is silent

Look at the App Audio item in the menu (“App Audio: zoom.us ♪ ▸”).

“♪” — the other participants’ audio is arriving.
“· No Sound Yet” — nothing is coming from the app right now. It is recorded as soon as someone speaks.
“· Not Running” — the chosen app isn’t running. Start the meeting app.
“· Not Available” — the app’s audio can’t be captured. Open the submenu for the reason and Check System Audio Recording….
If nothing arrives while the meeting app is playing, the “System audio recording” permission may be missing. Choose “Hearing nothing? Check System Audio Recording…” in the submenu and check that Minutes Ago is turned on.
During a meeting, the orange line “No sound from zoom.us for 30 s. Fine if nobody is talking” appears when the app has been quiet for 30 seconds.
The meeting app isn’t in the list — the submenu lists Meeting Apps that are open and Apps Using Audio. Join the meeting, then open the menu again.

See App Audio for details.

No transcript appears

Look at the second line of the menu.

“Preparing (language)…” — it is getting ready. Give it a moment.
“Downloading (language) · n%” — macOS is downloading the language model. Transcription starts when it finishes. Recording continues meanwhile.
It stays at “Nothing heard yet” while people are talking — check that Spoken Language ▸ in the menu matches the language spoken in the meeting. This is separate from the app’s language.
Lines stop appearing while sound keeps coming in — Minutes Ago restarts transcription by itself, first for that track, then for both. If the speech recognizer stops reading the audio and it does not come back, the orange line “Transcription has stopped (since …)” appears: choose Resume Transcription. While recording, Restart Transcription in the menu makes the live transcription of both tracks again at any time; recording goes on without a gap.
“Transcription unavailable: …” — check the reason shown.
Nothing transcribed at all — check that Transcribe while recording is on in Settings › Transcription.
Quiet speech is missing — outside meetings, Speech detection in Settings › Transcription decides what is transcribed. Standard skips what it hears as silence to save power, and quiet speech may be skipped too. Choose More Sensitive, or Always Transcribe (it uses more power). During meetings everything is transcribed while Always transcribe during meetings is on (the default). In the Recordings & Minutes window, the dot left of the Microphone and App Audio chips shows no sound, quiet (may not be transcribed), being transcribed, or off.
To fix the lines afterwards, transcribe the range again with another model (Transcription and Models).

Only final transcription results are saved, so a line appears a little after it is spoken.

Minutes don’t work

When Generate Minutes can’t run, the reason is shown (on the greyed button, and for providers in Settings › Meeting Minutes › Providers, with ⚠).

“Apple Intelligence is turned off. …” — turn it on in System Settings › Apple Intelligence & Siri, or make another provider the default in Settings › Meeting Minutes › Providers.
“The on-device model is not ready yet (it may still be downloading). …” — try again later.
“This Mac cannot run the on-device model. …” — on-device minutes can’t be written on this Mac. An external AI provider can still be used.
“Enter the API key in Settings › Meeting Minutes › Providers.” (or the model name, or the server URL) — the provider is chosen but not fully set up. Test Connection checks the key with one short request.
Apple Private Cloud Compute is coming later; it is shown greyed out in the current build.

Without Apple Intelligence, recording, transcription, export and speaker identification still work.

Speaker identification is slow

The first time, the model (about 200 MB) is downloaded (you can also download it beforehand in Settings › Models) and then loaded (about 23 seconds right after the download on the development Mac). After that, on the development Mac, a 5-minute range took about 2 seconds and a 90-minute range about 38 seconds, for one track. When the range has lines on both tracks (app audio and the microphone), each track is identified in turn, so it takes about twice as long.

After the first use, Minutes Ago’s memory use stays about 100 MB higher (it does not keep growing with repeated use).

A transcription model doesn’t download

Settings › Models shows the reason in the model’s row.

“Not enough free space on this Mac.” — free some space. The size is shown before downloading: Whisper large-v3-turbo 1.64 GB, Whisper large-v3-turbo (Compact) 645.7 MB, Parakeet TDT (Japanese) 619.1 MB, Cohere Transcribe (Beta) 2.19 GB.
“Hugging Face is busy. Try again later.” — the server is busy. Try again later.
“Couldn’t download the model. Check the network and try again.” — check the internet connection.
“Partly downloaded (…)” — the download was cancelled or Minutes Ago quit. Download… continues from there.
“Couldn’t optimize the model. Download it again or delete it and try again.” — delete the model with Delete… and download it again.

Models are downloaded only from Settings › Models, over HTTPS from Hugging Face. The speaker model is also downloaded the first time you identify speakers.

Optimizing or loading a model takes a long time

The first time a model is used on this Mac, it is optimized for it: “Optimizing for this Mac…”. Whisper on the GPU (the default) takes seconds, then it loads in about a second. Parakeet TDT (Japanese) takes about 2 minutes (“Optimizing for this Mac… (first time only, up to 15 minutes)”).
Cohere Transcribe (Beta) runs on the GPU and takes about 2 minutes to load (“Loading the model… (about 2 minutes)”) each time it is loaded, then about 6 minutes per hour of audio on each track. It stays loaded for 10 minutes after use, so another run within that time doesn’t wait for it to load.
While a model is being optimized in Settings › Models, it is greyed out in the Transcribe Again sheet and the menus (“Optimizing in Settings › Models”).
With Run Whisper on set to Neural Engine (under Advanced in Settings › Models), the first optimization can take 10 minutes or more and over 75 GB of free disk space (“Optimizing for this Mac… (can take 10 minutes or more on the Neural Engine)”, “Loading the model… (can take 10 minutes or more on the Neural Engine)”). Without enough free disk space it may not finish, and loading can take that long every time the app starts. If so, switch back to GPU.

A model is greyed out

In the Transcribe Again sheet, a saved recording’s “Transcribed with …” menu and the import sheet (Importing Audio Files), a model that can’t be used now is greyed out with the reason:

“Download in Settings › Models” — the model isn’t downloaded yet.
“Not available for this language” — the model doesn’t handle the spoken language. Parakeet TDT (Japanese) is for Japanese only. Cohere Transcribe (Beta) handles 14 languages (English, French, German, Spanish, Italian, Portuguese, Dutch, Polish, Greek, Arabic, Japanese, Chinese, Korean, Vietnamese); its Chinese is Mandarin, so it isn’t offered for Chinese (Hong Kong), which is Cantonese. Choose another spoken language in the sheet if the recording was in a different one, or another model.
“Optimizing in Settings › Models” — wait until the model’s row in Settings › Models reads Ready.
Transcribe Again… itself is greyed while another transcription runs (“Wait until the other transcription is done.”) or while the speakers or minutes of that recording are being made.

An audio file can’t be imported

Importing audio files needs Minutes Ago Pro (a license in the direct version); before you buy, every way in opens the purchase sheet (Importing Audio Files). When a file can’t be imported, the reason appears in red: in the import sheet while the audio is being prepared, otherwise at the bottom of the left pane of the Recordings & Minutes window, where “Recordings: N · size” usually is.

“This file can’t be read.” — it isn’t an audio or video file that macOS can read. Convert it to m4a, mp3 or wav, for example.
“This file has no audio track.” / “This file has no audio.” — a video without sound, or a file with no length.
“The model is not downloaded. Download it in Settings › Models.” or “This language can’t be transcribed. Choose another language.” — the model or spoken language chosen in the import sheet can’t be used. The recording made from the file is deleted; import it again with another choice.
“The audio couldn’t be written. Check the free space on this Mac.” (in the sheet) — the converted audio is kept as a saved recording, so it needs free space. Cancel closes the sheet.
Import is greyed out in the sheet with “Wait until the other transcription is done.” — only one transcription runs at a time. Wait for the other one (a range or a recording being transcribed again, or another import) to finish.
Dropping several files imports only the first one (“Only the first file is imported. Import the others one at a time.”).

Played lines are transcribed again as “Me”

When you play lines through the built-in speaker, the microphone can pick up the sound, and it is transcribed again as your own lines (“Me”). The playback bar then shows “Playing through the speaker may be transcribed again as “Me”. Use headphones.” Listen with headphones.

A pause sign in a circle — the meeting is paused. Nothing is recorded and the meeting’s clock is stopped until you choose Resume Meeting in the menu (or Resume in the meeting bar). It stays paused after Minutes Ago restarts.
A triangle — the microphone could not start (above).
Faint waveform bars — recording is paused (outside a meeting) or stopped. With a timed pause, the first line of the menu reads, for example, “Paused · Resumes at 14:35 · 12:03 left”.

The buffer shows slightly more than 90 minutes

With a 90-minute Keep the last setting, the first menu line can read, for example, “Recording · 1:31:30 kept”. Audio is kept in 30-second pieces and deleted piece by piece (every 15 seconds), so it can run slightly over. Nothing is wrong.

The recent audio was gone after a restart

By default, the audio that is not saved yet is kept in this Mac’s memory. When the app quits (Quit Minutes Ago in the menu, a restart or log out of the Mac, an update), it writes that audio to the SSD once and picks it up at the next launch. After a crash, a forced quit or a power loss it cannot do that, and the audio in memory is lost; the transcript for that time no longer appears in the Recordings & Minutes window either. Audio during a meeting is written to the SSD, so at most the last 30 seconds are lost. To lose at most 30 seconds at any time, choose the SSD in Settings › General › Live audio (it writes more to the SSD: about 1.6 GB a day with the microphone and app audio).

No screenshots are taken

The orange line “Screenshots are off – allow Screen Recording in System Settings, then restart Minutes Ago.” — choose Open Screen Recording Settings…, turn on Minutes Ago, and restart Minutes Ago.
Minutes Ago asks for the Screen Recording permission only when you start a meeting with screenshots from the menu or the window. A meeting started by a key from any app, an agent or Shortcuts starts without screenshots if the permission is missing (“No Screenshots (No Permission)”).
“No window of … to capture” — the chosen app has no window that can be captured. Open its window. For a browser, the front tab is captured.
Fewer screenshots than expected — a screenshot identical to the last one kept is not saved (Skip identical screenshots in Settings › Recording, on by default). A still screen no longer adds one every minute.

Add Screenshot is greyed out

Add Screenshot works only during a meeting with screenshots, while recording, with the Screen Recording permission, and not while the meeting is paused or being saved. A meeting started without Take screenshots of can’t add them; turn on screenshots when you start the meeting. “No meeting with screenshots is in progress.” means the same.

A key from any app doesn’t work

In Settings › Keyboard › Keys from Any App, check the row of the action.

“Not available (used by another app)” — another app has registered the same key. Record a different one.
“No key arrived. Another app may be using it.” while recording the key — another app takes that key first. Choose a different one.
The key must include ⌘ or ⌃ (with ⌘, also ⌃, ⌥ or ⇧). Keys used inside Minutes Ago are refused (“Already used for …”).
While Minutes Ago shows a confirmation (such as the free try question), keys only beep.

An agent can’t connect

In Settings › Agents, check the status: “Ready for MCP clients on this Mac” in the App Store version, “Listening on 127.0.0.1:48151” in the direct version. If it reads “Off”, turn it on; if “Unavailable: …”, check the reason shown.
App Store version: every request gets “Agent access is off. Turn on “Allow AI agents” in Minutes Ago › Settings › Agents.“ — turn on agent access; the next request goes through. If you moved the app after registering, copy the configuration again and re-register (the registered path points to the old place).
Direct version: after Regenerate token, existing configurations get 403. Copy the configuration again and re-register.
Direct version: if the port has moved to 48152 or later, it no longer matches the registered URL (48151). Copy the configuration again.
Direct version: use http://127.0.0.1:<port>/mcp or http://localhost:<port>/mcp. Other host names, and other machines, can’t connect (403).
export_audio fails — no export folder has been chosen (Settings › Library › Export folder).
generate_minutes fails — the default provider sends text off the Mac, and “Don’t show again” hasn’t been chosen for that destination.
add_screenshot or add_marker fails — no meeting with screenshots is in progress, Minutes Ago is not recording, or the meeting is paused.
pause_meeting or resume_meeting fails — no meeting is in progress, or it is already paused (or not paused).
Speaker identification or minutes fail — the client’s tool timeout may be shorter than the processing time (minutes for the last 30 minutes took about 100 seconds on the development Mac).

See Agents and Shortcuts for details.

`shortcuts run` doesn’t return

The first time each shortcut outputs a result, the Shortcuts app asks whether it may. shortcuts run doesn’t return until that is answered.

Passing the name of an app action (such as Get Transcript with Speakers) to shortcuts run doesn’t work. Use the bundled shortcuts (Install shortcuts… in Settings › Agents), or make your own shortcut in the Shortcuts app and run it by its name.

Changing the app’s language

Choose Language & Region… under Advanced in Settings › General. It opens Language & Region in System Settings, where you can choose a language for Minutes Ago under Applications. Menus and windows use that language after Minutes Ago restarts. (If System Settings can’t be opened there, choose System, 日本語 or English instead and click Restart Now.) Recording pauses for a moment while it restarts; live audio is kept.

The language spoken in meetings is a separate setting (Spoken Language ▸ in the menu). Minutes are written in the app’s language.

Frequently asked questions

Are the other participants of an online meeting transcribed too? Can I keep them separate from the microphone?

Yes. Choose your meeting app in App Audio ▸ in the menu and its sound is recorded as a second track, so the other participants are transcribed too. On export, the audio comes as two files (… mic.m4a and … app.m4a) and the transcript as one, with you as “Me” and the others as “Others” (after Identify Speakers, the others are S1, S2, … and you M1). If you listen on speakers, the app’s sound may reach the microphone and the same words can appear on both tracks. To record only the other participants, set the microphone to None.

Does a bot join the meeting?

No. The sound is picked up on your Mac, so nobody is invited and no participant is added. Tell the other participants yourself that you are recording, and that the screen is being captured if you take screenshots.

How can I tell it is recording?

The waveform bars of the menu bar icon are solid, and macOS shows its microphone indicator. The first line of the menu reads “Recording · … kept” (during a meeting, its name and elapsed time). While paused, the bars turn faint; while a meeting is paused, the icon is a pause sign in a circle. While screenshots are being taken, the icon shows a red dot. A triangle means the microphone could not start.

What is sent off this Mac? What does the developer receive?

Only the connections in the table: seven for the App Store version, eight for the direct version. Even when you write minutes with an external AI provider, only transcript text is sent (and screenshots, only with Include screenshots on), never audio. The developer receives no audio, transcripts, minutes, settings or usage data. Crash reports may reach us through Apple only if you agreed to “Share with app developers” in macOS.

How accurate is transcription of Japanese meetings?

Transcription uses Apple’s on-device speech recognition (SpeechAnalyzer). On a synthetic meeting (two speakers, 2 minutes 7 seconds), the character error rate (CER) through the app’s own path was 13.7–14.9%. Accuracy in real meetings has not been measured yet. Proper nouns it tends to mishear can be fixed with the replacement dictionary. The conditions and results are in About accuracy. A range or a saved recording can also be transcribed again on this Mac with Whisper large-v3-turbo, Parakeet TDT (Japanese) or Cohere Transcribe (Beta); their accuracy in real meetings has not been measured either.

Does it work for meetings in languages other than Japanese?

Transcription can use any language this Mac’s speech recognition supports (Spoken Language ▸ in the menu; up to five languages can be pinned there). A meeting can also use its own language, for that meeting only, through a preset or an agent. Only the Japanese synthetic meeting has been measured, though. If a meeting was recorded in the wrong language, a range or a saved recording can be transcribed again in another language, also with another model (Transcribe Again…). Parakeet TDT (Japanese) is for Japanese only, and Cohere Transcribe (Beta) handles 14 languages, including English and Japanese; a model that doesn’t handle the spoken language is greyed out. Minutes are written in the app’s language. If Apple Intelligence on this Mac can’t write that language, they are written in English (Changing the prompt).

Which macOS permissions does it ask for? Can I withdraw them?

The microphone; system audio recording only when you use App Audio; and Screen Recording only when you start a meeting with screenshots. You can withdraw any of them at any time in System Settings › Privacy & Security.

Can minutes be written by an outside LLM?

Yes. In Settings › Meeting Minutes › Providers, add OpenAI, Anthropic or Google Gemini with your own API key and a model name, or an OpenAI-compatible server (such as Ollama or LM Studio) with its URL, and make it the default (or pick it for one time with Minutes Options… in Generate Minutes ▾). API keys are stored in the Keychain, one per provider. Only the transcript text is sent, never audio, and the app names the destination and asks before sending. Each generation with a provider outside this Mac uses a free try until you get Pro.

Can I transcribe again with a different model?

Yes, on this Mac, free. Select a range in the live pane or open a saved recording, choose ⋯ › Transcribe Again…, and pick the Model: Apple (On Device), Whisper large-v3-turbo, Whisper large-v3-turbo (Compact), Parakeet TDT (Japanese) or Cohere Transcribe (Beta). The models other than Apple’s are downloaded once in Settings › Models. No audio is sent. For 8 seconds afterwards, Undo brings the previous lines back. A saved recording keeps one transcript per model: choose another model in its “Transcribed with …” menu, and switch between them at any time. Live transcription while recording stays with Apple. See Transcription and Models.

Can I listen to what was said?

Yes. Every line in the live pane and in saved recordings has a ▶ (or right-click › Play from Here). The playback bar under the list has a seek bar, a speed slider from 0.5× to 3× that keeps the pitch, and Follow. Space plays and pauses when the list has the focus. Your microphone and the app audio play together. In the live pane you can go back as far as the audio is kept, and playback stops 30 seconds behind now (“Caught up with live”). Use headphones: through the built-in speaker, the microphone can pick the sound up again. Playing is free. See Playing lines.

Can I fix or delete a line of the transcript?

Yes. Double-click a line (or right-click › Edit Text) to fix it in place; Return confirms and Esc cancels. Delete Line, or select lines and press ⌫, removes them (one line at once with Undo for 8 seconds, more lines after a question). Only the transcript changes; the audio is kept, so Transcribe Again… brings deleted lines back from it (and replaces edited ones). See Editing and deleting lines.

Can I pause a meeting?

Yes. Choose Pause in the meeting bar or Pause Meeting in the menu (also a key from any app, the Shortcuts app or an agent). Nothing is recorded and the meeting’s clock stops until you resume it, and it stays paused after Minutes Ago restarts. The saved recording has “Meeting paused” and “Meeting resumed” markers. Outside meetings, Pause ▸ in the menu and Pause ▾ in the live pane pause recording until you resume, for 5, 10 or 30 minutes, 1 hour, or a time you set (Custom…, up to 480 minutes). Pausing is free.

Can I mark the moments I want to come back to during a meeting?

Yes. Add a marker while recording, and it appears in the Recordings & Minutes window between the lines, with its time and name. You can add one with Add Marker in the menu (hold ⌥ to name it), a key from any app you set in Settings › Keyboard (none by default), the Add Marker action in the Shortcuts app, an agent (add_marker), or Add Marker in the window. Markers can be added only while recording (Add Marker Here on a line adds one at that earlier time). Right-click a marker to make a range such as To Next Marker or From Here to Now, or choose Since the Last Marker in the range ▾. Markers in the range go into copied text, exported files and the minutes ({{markers}}), unless you turn off Include Markers. Markers are free. See Markers.

Can I search what was said earlier?

Yes. Press ⌘F in the Recordings & Minutes window for a search bar that searches the whole transcript of live audio, or of the saved recording you opened. Upper and lower case, full-width and half-width characters, and hiragana and katakana are treated as the same. Return or ⌘G goes to the next match, ⇧Return or ⇧⌘G to the previous one. Moving to a match does not change the range. Search is free, and it also works in recordings imported from audio files. See Search.

Can I keep a meeting beyond the retention period?

Yes. Choose Start Meeting… in the menu when a meeting begins (a name is optional) and Stop & Save Meeting when it ends. While a meeting runs, nothing from its start is deleted by the retention period (up to 24 hours). The saved meeting stays on this Mac as a recording, in the list on the left of the Recordings & Minutes window, until you delete it; its minutes are kept with it. You can also save any range with Save to Library…. Starting a meeting is free; saving is counted like exporting (free within the last 30 minutes without speaker names, app audio included). Reading, copying and exporting a saved recording are free. See Meetings and Saved recordings.

Can it capture the screen of the video call?

Yes, during a meeting. In Start Meeting…, turn on Take screenshots of, choose the app and an interval (3 to 60 seconds). Only that app’s window is captured. The screenshots appear above the transcript and follow it as you scroll. Add Screenshot (in the meeting bar, the menu, or a key from any app) keeps the screen right now. With Include screenshots on in Settings › Meeting Minutes, the minutes can use the slides and figures on screen. Screenshots are free. See Screenshots.

What files are exported? Can I change the format of the minutes?

Export writes audio as .m4a and transcripts as .txt, .md and .json. Export Transcript… in Copy Text ▾ saves or copies the transcript as WebVTT or SRT subtitles, Markdown (paragraphs, list or table) or text, with or without speaker names and times (Export Transcript). Minutes become a Markdown file (.md) only when you save them (a saved recording keeps its minutes in its folder). You can change how the minutes are written with minutes presets in Settings › Meeting Minutes. Markers in the range also go into the .txt, .md and .json (unless you turn off Include Markers). A recording imported from an audio file exports the same way (The imported recording).

Can I transcribe an audio file of a meeting I didn’t record?

Yes, with Minutes Ago Pro (a license in the direct version). Drop an audio or video file (such as m4a, mp3, wav, mp4 or mov) on the Recordings & Minutes window, or choose Import Audio File… (File menu, ⌘O, or the menu bar icon’s menu), or the + at the bottom of the left pane. Choose the spoken language and the model, and click Import. The file becomes a saved recording with the file’s name and date and is transcribed on this Mac; after that it works like any saved recording (play, edit, identify and name speakers, minutes, Transcribe Again, export). The original file is not changed. This feature is not covered by the 5 free tries and can’t be used before you buy. See Importing Audio Files.

Does uninstalling remove everything?

Moving the app to the Trash may leave its container (~/Library/Containers/com.kurokoto.minutesago/) behind. First choose Pause ▸ Erase Live Audio… in the menu and Delete All Saved Recordings… in Settings › Library, and delete the container too if needed. Exported files stay in your folder (Uninstalling).

How much does it cost? Is it a one-time purchase?

The app is a free download from the Mac App Store. Minutes Ago Pro is a one-time in-app purchase with no subscription: ¥4,800 in the Japanese App Store (tax included), and ¥3,000 at the launch beta price. The beta price’s end date will be announced later, and beta buyers also get free updates through 1.x. Prices differ by country or region; the price shown in the App Store applies. The direct-download version’s license is sold separately, in US dollars ($19.99 at the launch beta price, then $29.99). See also Pricing.

What is free?

Saving, copying and minutes on this Mac for the last 30 minutes are free forever, app audio included, with no limit on how many times. Older audio, speaker names, providers outside this Mac and agents can be tried 5 times, and Minutes Ago Pro removes the limit. Importing audio files is for Minutes Ago Pro only and is not available with free tries.

Always free, whether or not you buy:

Recording never stops, and retention follows your setting (30 minutes to 24 hours). Pause (including timed pauses and pausing a meeting), Erase Live Audio… and where live audio is kept.
Viewing everything kept in the Recordings & Minutes window, playing lines, identifying speakers and naming them, and editing and deleting lines and screenshots.
Transcribing a range or a saved recording again with any model on this Mac, keeping a transcript for each model, and downloading and deleting models in Settings › Models.
Searching the transcript, and markers (adding, naming and deleting them, and making ranges from them).
The replacement dictionary, presets, and choosing App Audio and the microphone.
Save the Last ▸, Copy Text (and Export Transcript…), Export, Save to Library… and Generate Minutes (on this Mac) for the last 30 minutes, as long as the output has no speaker names (app audio included).
Meetings (starting, discarding) and screenshots, including Add Screenshot.
Saved recordings: reading, renaming, deleting, identifying speakers, Transcribe Again…, and copying, exporting and minutes on this Mac (they were paid for when saved). Only speaker names added after saving use one free try, once.

Minutes Ago Pro unlocks:

Taking out anything older than 30 minutes.
Exports, copies, saved recordings and minutes with speaker names.
Minutes from providers outside this Mac (OpenAI, Anthropic, Google Gemini, an OpenAI-compatible server; Apple Private Cloud Compute is coming later).
Taking things out through agents (buffer status, the last 30 minutes of transcript, starting meetings, listing and reading saved recordings, markers and screenshots stay free).
Importing audio files, with any model. Not covered by free tries, so it can’t be used before you buy. In the direct version, it works once you activate a license. Once imported, the recording is a saved recording, and copying, exporting and minutes on this Mac for it are free.

What do “(Free Try)” and “(Pro)” mean?

Before you buy, a button or menu item shows “(Free Try)” only when using it now takes a free try (for example 60 Minutes in Save the Last ▸, or Export for a range with speaker names). When no free tries are left it shows “(Pro)” (“(License)” in the direct version). Before a free try is used, the app tells you why and how many are left, and offers a free alternative where there is one (such as Only the Last 30 Minutes (Free) or Without Speaker Names (Free)). With Pro there are no marks. The lock icons of earlier versions are gone. The “(Pro)” mark on Transcribe Audio File… is different: free tries don’t cover it, and choosing it opens the purchase sheet.

What counts as one free try?

One save, export, copy, minutes generation or agent request that goes beyond the free range: older than 30 minutes, with speaker names, or with a provider outside this Mac (app audio within the last 30 minutes is free). Saving a meeting or a range as a recording counts like an export. It counts once however many of these it includes, and only when it succeeds; cancelling, or a failed export or generation, does not use one. The app shows how many are left before you use one. Viewing, playing lines, searching, markers, meetings (including pausing them), screenshots, identifying speakers (again), naming them, editing and deleting lines, Transcribe Again… with any model, switching between a recording’s transcripts, downloading models, and copying or saving minutes you already generated do not count. A saved recording is paid for once, when it is saved; later copies and exports of it are free, except speaker names added after saving (one free try, once). You have 5 in total; when they run out, the free range stays as it is. Transcribing audio files never counts (free tries don’t cover it; it needs Minutes Ago Pro or, in the direct version, a license).

Can I use agent access for free?

Yes, within the same limits as the app. Turning it on, get_buffer_status, get_transcript for the last 30 minutes, app audio lines included (speaker names are left out), start_meeting, pause_meeting, resume_meeting, stop_meeting with discard, list_recordings, get_recording, add_marker and add_screenshot are free and do not count. Older transcripts, transcripts with speakers (get_speaker_transcript), export_audio, generate_minutes and saving a recording (stop_meeting, save_recording) each use a free try until you get Pro. The response tells the agent how many are left. With none left, or for a range the free version can’t take out, the tool returns an error instead of quietly shortening the range. The same applies to the Shortcuts actions. See What agents can use for free.

Is there a command-line (CLI) or direct-download version?

The direct-download version, distributed from this website (a Developer ID signed and notarized DMG, with license keys from Polar), includes the command-line tool minutesago. How to use it is in Command-line tool (direct version). For how to get the direct version and whether it is available yet, see Pricing. Its license is sold separately from the App Store version, in US dollars ($19.99 at the launch beta price, then $29.99). Purchases do not carry over between the App Store version and the direct-download version. To use the App Store version from Terminal, use Shortcuts (shortcuts run) or agent access (MCP).

Who do I contact?

Email [email protected].