Once a recording is transcribed, ThinkScribe turns it into something you can act on: a meeting summary with action items, formal minutes, an executive brief, a mind map, or the answer to a specific question. It is written by ThinkScribe AI, which runs on your device, or by Apple Intelligence (provided by Apple) on devices that support it. Your transcripts are never sent to us, and summary bullets point back to the lines they came from.
How to generate a meeting summary from a transcript
- Open a transcript. On iPhone and iPad its tabs are Transcript, Summary and Chat; on Mac the Workspace has Transcript, Summary, Action Items, Chat and Mind Map.
- Choose Summarize (on Mac you can also use View > Summarize, Shift-Command-S).
- For a different shape, use the chips: Summarize, Meeting Minutes, Executive Brief and Mind Map. The summary includes an Action Items list.
- Tap a citation on a summary bullet to jump to the transcript line it came from, and play the audio there.
- Not quite right? Edit it or try again.
The result is stored with the transcript, so it is there the next time you open it, and it syncs with the transcript if you turned on iCloud Sync.
Which engine writes the results is set under Settings > AI Assistant on iPhone and Settings > AI & Models on Mac: Apple Intelligence or ThinkScribe AI, a one-time download.
Action items, minutes, briefs and mind maps
- Summarize: what was discussed, with the main points and an Action Items list of who agreed to do what.
- Meeting Minutes: a structured record with topics, decisions and next steps. Meeting minutes from a recording walks through a full example.
- Executive Brief: the short version for someone who was not in the room.
- Mind Map: the topics and how they relate, as a map you can switch to an outline and copy as Markdown or plain text. Its nodes link back to the moment in the recording. On Mac it is a Workspace tab; on iPhone it needs a recent model with more memory. Turn a recording into a mind map shows the steps.
On Mac you can also save your own quick actions, so an instruction you type every week (a client follow-up email, a blog outline) becomes one click on every transcript. Mac also has a Clean Transcript action. On iPhone the transcript already shows cleaned text by default, and ... > Show Original shows exactly what was said.
Ask questions about a recording in Chat
Asking the recording is for questions the summary does not cover:
- Open the transcript and start a question (on Mac, View > Chat About This Recording, Option-Command-C).
- Type into "Ask about this recording...", or tap a suggestion such as "What are the main topics discussed?", "What decisions were made?" or "Summarize this in one sentence".
- Read the answer, written from the transcript, and ask follow-ups.
On Mac, Chat answers show sources you can click to jump to the moment. On iPhone, Chat answers are plain text; for answers that point to the moment, use the Summary results and mind map nodes. Chat with a recording goes deeper, with questions worth asking.
To ask across several recordings, put them in a collection and ask the collection: answers come from all of its recordings.
Long recordings
A two-hour workshop is a long document for any model to read, so expect a very long recording to take a little longer to summarize than a short call.
What a language model on your own device does well
A model that runs on a laptop or a phone is smaller than the ones behind the large cloud assistants, and the gap shows up in predictable places rather than everywhere. It is strong at work that is mostly reading and reorganizing: pulling the decisions out of an hour of talk, listing who agreed to do what, condensing forty minutes into six sentences, noticing where one topic ended and the next began. That work needs comprehension, and everything it needs is in front of it.
It is weaker wherever the job needs something the transcript does not contain. It cannot tell you whether a plan is a good one. It does not know your company's history unless somebody said it out loud. It will not reliably count things across a long document. And it can only hold so much text at once: one minute of conversational speech is roughly 150 words, so an hour of meeting is already a long document and a half-day workshop is a very long one. Past that limit the transcript has to be summarized in pieces and the pieces combined, which is why a detail dropped early does not reappear at the end.
How to ask for a summary that is actually useful
The instruction matters more than most people expect. A bare "summarize this" produces the beige paragraph nobody reads. When the built-in chips are not enough, ask in Chat (or save a quick action on Mac) and include four things:
- Name the audience. Notes for someone who missed the call are a different document from notes for the person who ran it.
- Ask for a shape. Say how many bullets, which sections, and in what order. Decisions, then owners, then open questions beats prose every time.
- Say what to leave out. Scheduling, small talk, the ten minutes about the broken projector.
- Ask it to admit gaps. Telling the model to say when the transcript does not cover something is the cheapest guard there is against a confident invention.
Share the result
To send notes onward, export or share the transcript: iPhone, iPad and Mac export TXT, SRT, VTT, JSON, Markdown, PDF, Word and CSV, and the Mac adds Share to Notes. The export guide covers every format. For the wider workflow from recording to notes, see the complete voice to text guide.
Frequently asked questions
Can AI summarize a meeting recording?
Yes, in two steps. The recording is first turned into a transcript, and then a language model reads that transcript and writes the summary. The quality of the second step depends almost entirely on the first: clear audio with speakers told apart gives notes you can rely on, while a muddy recording produces confident nonsense.
How do I get action items out of a meeting transcript?
Ask for them separately from the summary, and ask for a shape: one line per item, in the form owner, task, deadline if one was stated. A general summary buries commitments in prose, while a dedicated pass lists them. Then check each item against the moment it came from before you send it to anyone.
Can I summarize a transcript without sending it to a cloud service?
Yes. Models small enough to run on a laptop or a recent phone handle summaries and action items well, because that job is mostly reading and reorganizing rather than world knowledge. ThinkScribe can use ThinkScribe AI, a one-time download that runs on your device, or Apple Intelligence (provided by Apple) on devices that support it. There is no account and no API key, and your transcripts are never sent to us.
How accurate are the summaries?
Good for meetings, lectures and interviews with clear audio and speakers told apart. The citations on summary bullets exist so you can check anything important in seconds rather than trusting a paragraph blindly, and Adjust or Retry fixes a result that missed the point.
