Speaker separation

Speaker separation is a transcription step that labels each line of a conversation by speaker role, such as prospect or rep, so what the prospect said stays distinct from what the rep said.

At your booth, each conversation mixes your rep’s pitch with a prospect’s needs. The prospect’s lines hold the pain points and product interest, and your rep’s lines hold the pitch and, often, the next step they promised.

Why does it matter for your lead records?

Structured extraction reads the transcript to fill in your lead fields. With labeled lines, it records a pain point in the prospect’s own words and traces each commitment to the rep who made it.

Without labels, your rep’s pitch can leak into the record as if the prospect had said it.

How does it differ from speaker identification?

Separation works within a single conversation and assigns roles: this line came from the prospect, that one from the rep. Speaker identification goes further and matches a voice to a specific known person.

For a clean record from conversation capture, role labels are all you need.

Where do you find it in FieldXM?

In the FieldXM mobile app, the live transcript runs on your rep’s phone, and speaker separation labels the prospect’s lines and your rep’s. It identifies no one.