From: allanh@vivaldi.net   
      
   On Mon, 04 Aug 2025 06:41:28 -0500, John C.  wrote:   
      
   > Allan Higdon wrote:   
   >>   
   >> This is a Speech to Text program that interests me.   
   >>   
   >> "Faster-whisper is an optimized implementation of OpenAI's Whisper model   
   >> for automatic speech recognition (ASR). It aims to provide faster   
   >> inference times compared to the original implementation while   
   >> maintaining accuracy. The project leverages CTranslate2, a fast   
   >> inference engine for Transformer models, to achieve improved performance."   
   >>   
   >> https://best-of-web.builder.io/library/SYSTRAN/faster-whisper   
   >>   
   >> "Standalone executables of Faster-Whisper &  Faster-Whisper-XXL for   
   >> those who don't want to bother with Python."   
   >>   
   >> Home Page   
   >> https://github.com/Purfview/whisper-standalone-win/   
   >>   
   >> Changelog   
   >> https://raw.githubusercontent.com/Purfview/whisper-standalone-win/main/   
   >> changelog.txt   
   >>   
   >> Downloads   
   >> https://github.com/Purfview/whisper-standalone-win/releases/download/   
   >> Faster-Whisper-XXL/Faster-Whisper-XXL_r245.4_windows.7z   
   >> https://github.com/Purfview/whisper-standalone-win/releases/download/   
   >> Faster-Whisper-XXL/Faster-Whisper-XXL_r192.3.4_windows.7z (Windows 7   
   >> Compatible)   
   >   
   > This does look interesting. Have you tried it out yet Allan?   
   >   
      
   I first tried it a couple of years ago.   
   The accuracy is not bad, but in my case, it takes hours to generate subtitles   
   for a 90-minute movie.   
   It's much better for those with a newer graphics card than mine.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)