Skip to content

Missing UI Overlay, Audio Cues, Hotkey Bug & AI Post-Processing Request #5

Description

@moj02090

Thank you for building this tool! I tested the software today for the first time. The dictation quality itself is excellent, but I encountered a few bugs and UX limitations during my testing. If you wanna compete with wispr flow or tools like handy

🐛 Bug Reports & UX Issues

  1. Ambiguous / Generic Audio Device Names
    Issue: The audio input device dropdown only displays generic Windows device categories (e.g., Headset Microphone, Echo Cancelling Speakerphone) without the specific hardware/model names.

Comparison: Standard Windows settings (and other apps) display the full device model beneath or next to the device name (e.g., Poly BT700, Yeti Stereo Microphone, or AMD Audio Device).

Impact: Users with multiple microphones or headsets cannot easily distinguish which physical device they are selecting.

  1. Missing Visual Overlay
    Issue: No visual overlay or on-screen indicator is displayed when recording.

Impact: There is no visual feedback to show whether the software is actively listening or processing speech.

  1. Hotkey Lock-Up / Repeating Character Issue
    Issue: After automatic punctuation triggers (inserting a period once a sentence ends), the active hotkey gets stuck in an infinite input loop.

Reproduction Steps:

Set the dictation hotkey to Ctrl + Y.

Dictate a sentence.

As soon as auto-punctuation inserts the trailing period, the letter y starts typing endlessly (yyyyyyyy...).

Root Cause: Auto-punctuation insertion seems to interfere with hotkey release detection.

💡 Feature Requests & Improvements

  1. Audio Cue on Recording Start
    Current Behavior: An audio chime only plays when stopping transcription.

Request: Add an audio tone when recording starts so users immediately know the software is listening—especially helpful when no overlay is visible.

  1. Advanced AI Post-Processing (Custom Prompts & OpenRouter)
    Request: Add an option to enable automated post-processing via custom AI providers (e.g., OpenRouter or direct API integrations).

Details: Allow users to define custom prompts (e.g., cleaning up filler words, reformatting, or restructuring text) that run automatically on the raw dictation before inserting it.

  1. Flexible Hotkey Configurations
    Toggle Mode: Support pressing a single hotkey once to activate recording and pressing it again to deactivate.

Dedicated Keys: Allow setting separate, dedicated hotkeys for Start Recording and Stop Recording.

Gerätename PC2FJW9Y
Prozessor AMD Ryzen 5 PRO 5650U with Radeon Graphics (2.30 GHz)
Installierter RAM 16,0 GB (14,8 GB verwendbar)
Systemtyp 64-Bit-Betriebssystem, x64-basierter Prozessor

Edition Windows 11 Business
Version 25H2
Installiert am ‎25.‎07.‎2026
Betriebssystembuild 26200.8893
Funktionspaket Windows Feature Experience Pack 1000.26100.334.0

Thanks Moritz

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions