I built Utter, a native macOS dictation app, and wanted to share it here to get feedback.

The motivation was that Apple Dictation works for short notes, but for longer or technical input it often produces a wall of text that takes more time to clean up than typing. I wanted dictation that could understand context and produce structured output directly.

Utter works system-wide: you press a global hotkey, speak, and the text is inserted directly at your cursor in any app. Before insertion, the dictated text can optionally be processed through a custom AI prompt (fully user-defined).

Some details:

- System-wide dictation (works in any text field)

- Custom prompts to turn speech into emails, structured Markdown, summaries, JSON, etc.

- Bring-your-own API keys (OpenAI, Anthropic, etc.), so it’s free to use

- Optional hosted models if you don’t want to manage keys

- Privacy-first: no accounts, no data retention, keys stored locally

- Searchable audio + transcript history

- iOS companion app with a custom keyboard, synced via iCloud

My main use cases are note-taking, email, and coding. For example, I often dictate rough thoughts while walking and have them converted into clean Markdown, or dictate rough prompts into terminal and turn them into structured prompts with file references.

Link: https://utter.to

I’d really appreciate feedback — especially from people who use dictation or voice input regularly, or who’ve tried building similar workflows.

Thanks!