Professional transcription. Entirely on your Mac.
Audio and video become text with word-level timestamps, pauses, emphasis and emotion in the voice. No internet, no account, and not a single byte leaves your machine.
macOS 26 or later · Apple silicon Mac · ~480 MB
Nothing leaves this Mac.
No cloud
Transcription, text refinement and emotion analysis all run on your Mac's chip. It works in airplane mode.
No account, no tracking
There is no sign-up, telemetry or analytics. The app does not know who you are.
Your files stay yours
Interviews, meetings, sensitive material: nothing is ever sent to any server.
Features
State-of-the-art transcription
Whisper models running on the Neural Engine, from fast to large-v3, with a timestamp on every word.
Emotion in the voice
Neural networks analyse the actual audio, not the text, and label the emotion of each utterance, with optional per-word analysis.
Tone, pauses and emphasis
Short, notable and long pauses; emphasised words; rising or falling intonation; speech rate in syllables per second.
Refinement with local AI
Punctuation and correction with Apple Intelligence, on-chip MLX models or Ollama, with a glossary to get your terms right. Timings never change.
Exports the way you need
SRT, VTT, TXT and a complete JSON with every word, timing, pause and emotion, ready for video editors and pipelines.
Handles (almost) anything
MP4, MOV, MP3, WAV, M4A, FLAC and even MKV, WEBM, AVI and OGG. Drag, drop, done.
The only transcription that hears the tone, not just the words.
Video editors, podcasters, researchers and journalists use Whisper PRO to find the right take by its emotion: the hesitation, the emphasis, the dramatic pause, straight from the exported JSON.
- 00:12
I never imagined we would pull this off.
joylong pauseemphasis on “never” - 00:18
It was... complicated at first, honestly.
hesitationnotable pausefalling intonation - 00:24
And then everything changed at once.
surprise4.1 syllables/srising intonation
How it works
- 1
Drag in your file
Drop the audio or video onto the app window.
- 2
Your Mac does the rest
Transcription and analysis run on your Apple chip.
- 3
Export SRT, VTT, TXT or JSON
Ready for your timeline or your pipeline.
Download Whisper PRO
Version 0.13.0 · ~480 MB
Requirements
- macOS 26 (Tahoe) or later
- Apple silicon Mac (M1 or newer)
On first launch, macOS asks for confirmation in System Settings → Privacy & Security, because the app is still going through Apple's notarisation process.