Enhancing Transcript Accuracy with Fireflies' Speaker Labeling (Full Transcript)

Speaker 1: So Fireflies, if there's good audio and then there's a clear separation of speakers when you're talking on the meeting, it'll label it as Speaker 1, Speaker 2, Speaker 3. Typically the fewer people in the meeting, the better it's able to recognize voices. Once the labels have been created and you have your transcript, you can actually go back and bulk edit the names really quickly. So for example, I know from listening to this and reading the transcript that Speaker 1 was Nick. So I'm just gonna go and type in Nick for Speaker 1 and what you'll see is all Speaker 1's have been updated to Nick and then Speaker 2 was Rowan. So I'm gonna update that to Rowan and all Speaker 2's have been updated to Rowan. If you are on the web version, there's more advanced editing capabilities. So for example, if Rowan was misattributed to Nick, I can highlight that in the edit and then I can assign a new speaker to it. So most people really don't need to go that deep, but if you really wanted to get it absolutely perfect with speaker recognition, you could do that. The other advantage of doing all this is that it's actually training Fireflies to get better at recognizing different speakers over time. So hopefully that helps answer your questions. Thanks.