Video Extraction Plus

소개

Video Extraction Plus extends a core server for video and audio transcription by integrating diverse automatic speech recognition (ASR) capabilities. It offers flexible options, allowing users to choose between local transcription with OpenAI's Whisper model or leverage the powerful online ASR services from ByteDance's JianYing (CapCut) and Bilibili's Bcut. Designed for extensibility, the tool provides a standardized architecture for easy integration of new ASR providers, robust caching, rate limiting, and comprehensive status management, making it a versatile solution for automating transcription workflows.

주요 기능

• Built-in rate limiting for online ASR services to prevent API overload
• Configurable ASR provider via YAML file or environment variables
• Caching mechanism for ASR results to improve efficiency
• Extendable architecture for integrating new ASR providers
• Multi-ASR provider support: Whisper (local), JianYing (online), Bcut (online)
• 4 GitHub stars

사용 사례

• Leveraging online ASR services from JianYing (CapCut) for efficient transcription.
• Integrating Bilibili's Bcut online ASR for content related to the Bilibili platform.
• Automating transcription of video and audio files using local Whisper models.