Skip to content

Bug: Dictation appends/merges system clipboard text into transcription #78

Description

@HumbleBee14

Bug
Whenever I complete a dictation, open-wispr pastes the newly transcribed Speech-to-Text (STT) data along with whatever string is currently saved in my system clipboard. It mixes my old clipboard history directly into the active application text box.

To Reproduce

  1. Copy any text snippet.
  2. Trigger open-wispr and speak a sentence.
  3. Stop recording.
  4. The output text box prints: [Old Clipboard Text] + [New Spoken Text]

Expected behavior
The application should only paste the newly transcribed voice recording and leave the previous clipboard content untouched, or restore it silently without combining the text items together. It wasn't happening earlier, I think this started happening after I quit and restarted the service after switching model from base to medium.

Configuration Context

  • Tried turning toggleMode both true and false, but the problem persists.

  • Current configuration file:
    {
    "hotkeys" : [
    {
    "keyCode" : 63,
    "modifiers" : [

    ]
    

    }
    ],
    "hotkey" : {
    "keyCode" : 63,
    "modifiers" : [

    ]
    },
    "modelSize" : "medium.en-q5_0",
    "toggleMode" : false, # (tested with true also)
    "spokenPunctuation" : false,
    "language" : "en"
    }

Desktop:

  • OS: macOS

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions