OpenSuperWhisper

OpenSuperWhisper is a macOS application that provides real-time audio transcription using the Whisper model. It offers a seamless way to record and transcribe audio with customizable settings and keyboard shortcuts.

Features

• Real-time audio recording and transcription
• Two transcription engines: Whisper (opens in a new tab) and Parakeet (opens in a new tab) — download models directly from the app
• ⌨ Global keyboard shortcuts — key combination or single modifier key (e.g. Left ⌘, Right ⌥, Fn)
• Mouse button trigger — bind the middle or an extra (thumb) mouse button to start/stop recording
• Hold-to-record mode — hold the shortcut, modifier key or mouse button to record, release to stop
• Drag & drop audio files for transcription with queue processing
• Microphone selection — switch between built-in, external, Bluetooth and iPhone (Apple Continuity) mics from the menu bar
• Support for multiple languages with auto-detection
• 🇯🇵🇨🇳🇰🇷 Asian language autocorrect (autocorrect (opens in a new tab))

Requirements

• macOS (Apple Silicon/ARM64)

Whisper Models

You can download Whisper model files (.bin) from the Whisper.cpp Hugging Face repository (opens in a new tab). Place the downloaded .bin files in the app's models directory. On first launch, the app will attempt to copy a default model automatically, but you can add more models manually.

Hebrew (ivrit.ai)

For Hebrew transcription, download the "Turbo V3 Hebrew" model from Settings → Model. It is ivrit.ai (opens in a new tab)'s Hebrew fine-tune of whisper-large-v3-turbo (whisper-large-v3-turbo-ggml (opens in a new tab)) — the same base model as the other "Turbo V3" entries, but tuned for Hebrew. Selecting it automatically sets the input language to Hebrew, which these models require to be set explicitly.