Supports the following features
- Transcription using OpenAI Whisper and FasterWhisper
- Alignment with WhisperX
- Diarization using Pyannote
- Associated analytics - listen:speak ratio, average talking speed, longest speaker monologue, switches per conversation, average pause time, keyword spotting
The workflow also contains performance tuning guide to speed up inference on Intel Hardware.