Dictly

Dictly is a speech-to-text dictation tool designed for macOS, iOS, and iPadOS, offering fully on-device transcription with a focus on privacy and instant processing. It transforms spoken words into polished, structured text in real time, without sending any data to external servers or collecting analytics. The tool emphasizes privacy by ensuring all recognition, formatting, and analytics are computed locally, allowing users to dictate anywhere, even in airplane mode or secure environments, with no network usage.

Key features include programmable Workflows that let users customize the processing pipeline, enabling live transformations such as filler removal and Markdown formatting as speech is transcribed. Dictly supports Per-App Profiles, which automatically apply specific formatting and behaviors depending on the application in use—for example, bullet lists in Notes or formatted paragraphs in Mail. Customization options extend to auto-submit, scrolling, animation, speech model, and language settings, all handled locally. Users can invoke a Quick Capture overlay from anywhere using a configurable shortcut, dictating live and inserting the resulting text via Type-out, Paste, Copy, or Auto-Submit into any field. The tool maintains a history of transcriptions, with search and export capabilities available, and supports custom dictionaries for names and terms.

Dictly is delivered as a Swift-native, lightweight app (under 5 MB download, approximately 50 MB RAM usage) for Apple devices, with identical offline operation across macOS, iOS, and iPadOS. It does not require or use an internet connection, verified by network inspection, and stores data locally using SwiftData. An optional private CloudKit sync for settings is planned. 99, which unlocks unlimited Workflows, dictionaries, profiles, and export options, as well as unlimited transcription history with search and sharing features.

Positioned as a professional-grade dictation solution for Apple platforms, Dictly offers users control, privacy, and customization over their speech-to-text experience, with every aspect of processing and storage remaining entirely on the user's device.