MacWhisper

MacWhisper is a native macOS transcription app for people working with interviews, lectures, meetings and other recorded audio. It runs speech recognition on your own Mac, so local transcription keeps audio on your device. It also offers cloud transcription through services such as OpenAI, ElevenLabs and Deepgram, which send audio off your machine.

You can record microphone or system audio, import audio and video files, and capture meetings in Zoom, Teams and other calling apps. The transcript editor links text to audio playback, lets you revise individual segments and supports searches across transcripts and speakers. Folders help keep recordings organized, and automatic filler-word removal can clean up spoken text.

For video work, MacWhisper exports SRT and VTT subtitles and plays video alongside synchronized captions. Document exports include DOCX, PDF and Markdown. Its .whisper format keeps the original audio and transcript edits together for sharing.

Model support includes OpenAI Whisper, Qwen3-ASR and NVIDIA Parakeet, with local speaker recognition powered by NVIDIA Nemotron 3 Diarization. Apple Silicon Macs are recommended, though the app also runs on Intel Macs. Metal GPU acceleration supports local processing; Medium and Large Whisper models call for more than 8 GB of RAM. Qwen3-ASR requires macOS 26 and an M-series chip.

Apple Intelligence and Shortcuts integrations connect transcription to other Mac tasks. The app is also available on iPhone and iPad.