{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/1a846522f6064baaa908b2e8eb4842bc\" frameborder=\"0\" width=\"1728\" height=\"1296\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1296,"width":1728,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1296,"thumbnail_width":1728,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/1a846522f6064baaa908b2e8eb4842bc-8c5350ed67784d55.gif","duration":137.9,"title":"Custom AI Analysis for Call Transcripts","description":"This Loom explains how to set up custom AI analysis points to extract qualitative and numerical data from call transcripts. It covers creating labels for fuzzy signals like intent, sentiment (for example frustrated callers or requests to speak to a human), and primary use case, with selection options such as high, medium, and low intent. It also shows numerical extraction examples like seat count, including categorization such as 200 plus. A sample qualification call demonstrates extracted outputs like inbound support interest, qualification status, based in the United States, and a 150 seat count."}