Documentation Index
Fetch the complete documentation index at: https://hyperwhisper.com/docs/llms.txt
Use this file to discover all available pages before exploring further.
Common questions
- How do I transcribe an audio file I have on disk? Right-click the tray icon and choose Transcribe Audio File. Supported formats are WAV, MP3, and M4A. Full walkthrough on the Transcribe a File page (Windows tab).
- Do I need to set up an API key? No. HyperWhisper Cloud works out of the box and a Pro license includes 5,000 credits. API keys are only needed if you want to BYOK — see Providers and API Keys.
- Does HyperWhisper Cloud add a markup? No, it’s priced at face value over the underlying provider. See Providers.
- Which local model should I download? Whisper Small is the best balance for most users. If you’re on ARM Windows or want maximum English speed, use Parakeet v2. Full hardware-by-hardware guidance on the Local Models page.
- Do local models run offline? Yes — once downloaded, audio never leaves your device.
- What if I don’t have a dedicated GPU? Both engines fall back to CPU automatically. Pick a Tiny or Base Whisper model for the best CPU experience.
A “No suggestions.” popup appears after dictation
This popup comes from Windows, not HyperWhisper or your editor. Windows 11 has a built-in hardware-keyboard text prediction feature that watches any text field and shows a small prediction bar above the cursor — including the “No suggestions.” message when it has nothing to predict. It triggers whenever new text lands in a focused editable field, so it can appear after HyperWhisper pastes a transcript the same way it would after you type manually.How to turn it off
This is a known quirk that affects VS Code and other editors regardless of how text is inserted — see microsoft/vscode#212357 for background.
