All Whisper models converted to all Ctranslate2 format and quantized for use with Faster-Whisper library.