Expand description
Enhanced VAD functionality with segment aggregation
This module provides advanced VAD features beyond the basic whisper.cpp implementation, inspired by faster-whisper’s optimizations. VAD is a preprocessing step that happens BEFORE transcription, not part of the transcription API itself.
Structs§
- Audio
Chunk - Audio chunk with metadata for transcription
- Chunk
Metadata - Enhanced
VadParams - Enhanced VAD parameters with aggregation settings
- Enhanced
VadParams Builder - Builder for enhanced VAD parameters
- Enhanced
Whisper VadProcessor - Enhanced VAD processor with segment aggregation