New in version 1.3 • Drop a file: drag a video or an audio recording onto Soliloquium – the pill says “Drop here”, and a moment later the text is on your clipboard. Voice message, voice memo, podcast, screen recording, lecture: anything with sound in it. Long files go through piece by piece, and they are polished and translated just like a dictation. • Double-click to minimise: a double click on the pill sends it up into the menu bar – as long as no recording is running. • Copied items come along: switch on “Include copied items” in Settings, and what you copy during a recording – a link, a passage, a file in the Finder – is attached below your dictation. A paperclip counts along; a click on it leaves the items out. The clipboard is only read while a recording runs, passwords from password managers are skipped, and everything stays on your Mac. • Mac screenshots (⇧⌘3, ⇧⌘4, ⇧⌘5) taken during a recording join it more reliably – and click when they have arrived.

New in version 1.3 • Drop a file: drag a video or an audio recording onto Soliloquium – the pill says “Drop here”, and a moment later the text is on your clipboard. Voice message, voice memo, podcast, screen recording, lecture: anything with sound in it. Long files go through piece by piece, and they are polished and translated just like a dictation. • Double-click to minimise: a double click on the pill sends it up into the menu bar – as long as no recording is running. • Copied items come along: switch on “Include copied items” in Settings, and what you copy during a recording – a link, a passage, a file in the Finder – is attached below your dictation. A paperclip counts along; a click on it leaves the items out. The clipboard is only read while a recording runs, passwords from password managers are skipped, and everything stays on your Mac. • Mac screenshots (⇧⌘3, ⇧⌘4, ⇧⌘5) taken during a recording join it more reliably – and click when they have arrived.

New in version 1.2 • Polish the transcript: Apple’s language model goes over the finished dictation once, on your Mac – filler sounds, stumbles and doubled words go, full stops and commas land where they belong, and a new paragraph begins where the subject changes. Nothing is shortened, added or answered, and nothing goes to the network for it. Requires macOS 27 with Apple Intelligence, is off by default and can be switched on in Settings – it also works behind GPT Transcribe. • Translate: at the very end, Apple’s translation puts the finished dictation into the language you choose, on your Mac – no network, no key. A click on the target language’s flag leaves the translation out for that one dictation. Requires macOS 26.4, is off by default and is switched on in Settings. • Shortcuts: four actions of our own for Apple’s Shortcuts app – start recording, stop recording, take a screenshot, send a dictation. You can also pick an existing shortcut that runs after every finished dictation and receives the copied text. • Open API interface: the paper plane hands a finished dictation, with its context and screenshot paths, to other apps on your Mac. Off by default. • Four new languages: Arabic, Hebrew, Malay and Slovenian – Arabic and Hebrew run right to left.

New in version 1.1 • Screenshot shortcut: captures the screen under the pointer and attaches it to your dictation – from any app. Set it in Settings. • Include Mac screenshots: screenshots taken with ⇧⌘3, ⇧⌘4 or ⇧⌘5 during a recording are attached automatically. Turn it on in Settings. • Your OpenAI API key is now stored encrypted in the Apple Keychain. • Fixed: recording sometimes failed to start when the chosen microphone used a different sample rate than the default input (for example with AirPods).