Replies: 1 comment
-
I think this would be an awesome feature, maybe even essential. It would be nice if you could select the transformation before you start with a shortcut and then as suggested get a menu to select the transformation. A bonus feature would be to have another shortcut for applying transformations on the last recording, so that you select the transformation and it transforms the last recording/transcript with the selected transformation. So when you selected the wrong transformation or you forgot to switch transformations, it's still usable and you can still re-transform it without having to record the message or the text again (without having to open the window of Whispering). And as suggested, maybe in a future version it would be even greater if it would be linked to the application you're using. For example, when using Outlook or Discord, use different transformations. |
Beta Was this translation helpful? Give feedback.
-
Problem Statement
Currently, Whispering transcribes voice input, but different use cases require different text transformations. Users often switch between various applications (e.g., task managers, ChatGPT prompts, casual chats) where different text structures and formatting styles are needed. Manually adjusting formatting after transcription is inefficient and inconsistent.
Additionally, users who pay for OpenAI services (e.g., Whisper and ChatGPT) may want to control and optimize their usage to avoid unnecessary processing costs.
Proposed Solution
Introduce a customizable transformation selection system that allows users to apply different formatting options on-the-fly using keyboard shortcuts.
Quick Selection via Modal Window:
A new customizable keyboard shortcut (e.g., Ctrl + Shift + T) opens a small modal window displaying available transformation options.
Users can navigate with arrow keys to select the desired transformation before recording automatically starts.
Default Transformation Per App:
Users can assign default transformations for specific applications (e.g., structured formatting for task managers, raw input for ChatGPT).
If a default transformation for currently opened app exists, the standard shortcut applies it instantly without requiring manual selection.
Re-transform
The modal window should also include an option to reapply a different transformation to the most recent transcription. This allows users to quickly correct formatting if the wrong transformation was initially selected. The feature should be keyboard-accessible (e.g., via a checkbox or a dedicated shortcut) to ensure seamless workflow adjustments without requiring manual re-recording.
Would love to hear thoughts on feasibility and implementation!
Beta Was this translation helpful? Give feedback.
All reactions