Description
Speech recordings can be transcribed with Whisper models accelerated by CTranslate2. Developers and researchers use this library for subtitles, meeting notes, media indexing, and speech experiments. Audio may contain voices, names, or private conversations, so consent and retention policies matter.