StreamWave v0.11.2

Transcript setup

• The Transcript panel helps you set up whisper.cpp instead of the old "whisper.cpp not found" message. That message also suggested winget install whisper.cpp, a package that doesn't exist. The panel now:

  • explains what whisper-cli is and what "on PATH" means;
  • shows the right install command for your system with a Copy button: scoop install whisper-cpp on Windows, brew install whisper.cpp on Mac, pacman/apt on distros that package it, or building from source;
  • has Open terminal: on Linux the command is typed in for you to run; on Windows and Mac it is copied for pasting. Nothing runs until you press Enter.

• Set it up for me (Windows and Linux): downloads the official whisper.cpp build from the ggml-org/whisper.cpp GitHub releases and a speech model from Hugging Face. A confirm dialog lists the files, where they come from and their sizes, and lets you pick the model. You get progress and Cancel. Files are checked by sha256 and size, then installed into StreamWave's own folder and used from there, so no PATH change is needed. On Windows you can optionally add that folder to your user PATH. Remove undoes everything.

• I already have it…: choose whisper-cli yourself (checked before it's saved), then Check again.

• Safer lookup. In order:

  1. the file you chose;
  2. StreamWave's copy;
  3. whisper-cli next to the app;
  4. whisper-cli on PATH.

  A generic main or bare whisper program is never run. Downloads only go over HTTPS to GitHub and Hugging Face, and redirects are checked too.