Speaker Detective: How Machines Learn Who's Speaking

Dive into speaker diarization—the tech that splits audio by speaker and labels who spoke when. We’ll explore voice prints, voice activity detection, and clustering; how accuracy is measured with diarization error rate and the newer conversational DR; and why ground-truth data matters. Real-world uses range from meeting transcripts to call-center analysis and AI assistants that summarize conversations.

Note: This podcast was AI-generated, and sometimes AI can make mistakes. Please double-check any critical information.

Sponsored by Embersilk LLC