Comparison
MacWhisper and Minutes are on the same side of the line that matters most: transcription runs locally on your Mac. MacWhisper can offer Whisper and Parakeet there; Minutes offers local engines according to platform, build, and model availability. The larger difference is the shape of the job. MacWhisper is the best drag-and-drop file transcriber on macOS — files in, transcripts and subtitles out. Minutes is a conversation memory layer — meetings and memos in, a growing structured archive out, one your AI agents can query. Respect where it's due; this is a comparison between two local-first tools, and many people legitimately want the other one.
Quick verdict
Choose MacWhisper if your job is transcribing files — interviews, podcasts, videos, YouTube links — and you want the most polished Mac GUI for it, with subtitle export and a one-time price.
Choose Minutes if your job is remembering conversations — recording meetings and memos into a private, diarized, searchable archive that Claude and other agents can use — and you want it open source and free.
| Category | MacWhisper | Minutes |
|---|---|---|
| Best for | Transcribing audio/video files on a Mac with a polished GUI | On-device conversation memory: meetings, memos, and dictation your agents can query |
| Core job | File in, transcript out — batch jobs, subtitles, YouTube and media-file URLs | Capture conversations, diarize them, keep a searchable structured record |
| Where transcription runs | On-device (Whisper, Parakeet, and other local models) | On-device, using the engines available for your platform and build |
| Optional cloud AI | BYO API keys for summaries/chat (OpenAI, Anthropic, others) — or fully local via Ollama/LM Studio | Optional and explicit: Claude via MCP, a local LLM (Ollama), or BYO-key cloud providers — off by default |
| Durable output | Exports: txt, srt, vtt, md, pdf, docx — per file | Markdown archive with YAML frontmatter, action items, decisions — per conversation, organized over time |
| Speaker handling | Automatic speaker recognition (Pro) | Diarization plus confidence-aware attribution that learns real names |
| Agent / MCP surface | CLI control and workflow automations; no MCP server we could find | MCP server (34 tools), CLI, SDK, Claude Code plugin over your local files |
| Open source | No | Yes, MIT |
| Platforms | macOS (14+ for the App Store build) and iOS; no Windows or Linux | macOS and Windows desktop apps; Linux CLI |
| Pricing | Free and paid options; direct and App Store pricing differ | Open source and free to run yourself |
Where MacWhisper wins
Where Minutes wins
The overlap is real: both record meetings and transcribe locally. MacWhisper offers Whisper and Parakeet engines; Minutes offers local engines according to platform, build, and model availability. The deeper divergence is what happens after transcription. MacWhisper's output is a document you export and move somewhere; its center of gravity is the file. Minutes' output is an entry in a corpus — ~/meetings accumulates, search spans months, and MCP tools answer questions like 'what did we decide about pricing in April' across everything.
A fair test: open your transcription tool's output folder. If it's a pile of exports you rarely revisit, either tool works and MacWhisper is more polished. If you wish that pile were a queryable memory, that wish is the entire reason Minutes exists.
Pick MacWhisper for file work: interviews, podcast episodes, subtitle jobs, transcribing someone else's videos. It is the best tool on macOS for exactly that, with straightforward one-time pricing.
Pick Minutes for conversation memory: your own meetings and ideas, captured continuously into structured local files your agents read straight from disk — no vendor sync, no account.
Plenty of people should own both — they're neighbors, not rivals: one optimizes the transcript, the other optimizes the archive.
Minutes is not the right first choice if your work is transcribing files you receive — it's built around capturing live conversations, not batch-processing media libraries or producing subtitles.
It's also not the fit if you want an iOS-first experience or a polished GUI for one-off transcription jobs; MacWhisper is simply better at those today.
Release spot check, October 7, 2026: reviewed the linked official product pages for current workflow, provider, platform, and pricing claims. This is not a new hands-on benchmark. Minutes captures, transcribes, and stores conversation records locally. Engine availability depends on your platform, build, and installed models. If you choose a cloud summarizer or connect a cloud assistant, authorized meeting context can reach that provider. File sync and backups you configure are separate data boundaries.
This is a fit-based comparison between two local-first tools, reviewed on 2026-07-18 against MacWhisper's official site and App Store listing, linked below. MacWhisper's local-by-default transcription, Whisper/Parakeet engine support, Pro feature list, and separate direct and App Store editions are drawn from its own pages.
The Minutes side is grounded in its public docs and open-source repository. Both tools' privacy claims are architecture-level and, in Minutes' case, verifiable in source.
Next step