WhisperPlus is a Python library that wraps OpenAI’s Whisper model for speech-to-text transcription. It provides a SpeechToTextPipeline with support for 4-bit quantization via HuggingFace Transformers and the HQQ library. The library includes utilities to download audio from YouTube URLs and transcribe them. It integrates with HuggingFace Spaces and exposes models on the HuggingFace Model Hub.