# Long Audio Chunk Processing

Whisper is a pre-trained automatic speech recognition (ASR) and speech translation model, trained on 680,000 hours of annotated data with strong generalization capabilities.