You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
whispr is a transcription pipeline built around OpenAI's Whisper model. It supports four backends (three interchangeable general-purpose ones plus an optional Mega-ASR specialist for degraded audio), parallel chunk processing, automatic fallback, and multiple output formats.