Set up local meeting summaries in Soufflé

Summaries are an optional step after transcription. A supported Apple Intelligence configuration or an Ollama instance can turn the transcript into a summary, decisions, action items and open questions.

Updated

Start here

  1. Open Settings → AI and check which provider is available. Apple Intelligence requires macOS 26 or later and availability on your Mac; transcription does not require it.
  2. For Ollama, install the server and a chat model, then select that model in Soufflé. Keep the server URL on localhost to process the transcript on this Mac.
  3. Open a recorded meeting and generate its summary. Review decisions and action items before using them; you can edit the summary templates in Settings → AI.

Summaries

Summaries are generated on-device: by Apple Intelligence on macOS 26 or newer, with nothing to install, and otherwise by a local Ollama. Alongside the prose summary, Soufflé extracts decisions, action items with owners, and open questions.

Install and start Ollama, then download a model:

brew install ollama
brew services start ollama
ollama pull qwen2.5:7b

Then, in Settings → AI, leave the server at http://localhost:11434 and pick the model. Any general-purpose chat model works; qwen2.5:7b is the one Soufflé recommends. The same provider is what dictation polish uses.

With Apple Intelligence or Ollama on localhost, text is processed on your Mac. If you change the address to a remote Ollama server, Soufflé sends it the text needed for summaries or dictation polish. Choose that server accordingly.