Fostr automatically identifies who is speaking during meetings so your transcripts, summaries, and Actions reflect accurate insight, whether the meeting is fully remote, in person, or hybrid. This guide explains what attribution is, why it matters, and how to configure your audio setup for the best results.
Speaker attribution is Fostr's ability to link spoken words to the person who said them. During a meeting, Fostr primarily relies on speaker attribution data and transcripts provided by your integrated meeting platform (Microsoft Teams, Zoom, or Google Meet). When a platform-provided transcript is available, Fostr ingests that transcript directly and preserves the speaker labels supplied by the provider. Fostr applies its own local voice separation model only when a transcript is not available.
Accurate attribution keeps your meeting content useful and clear.
When Fostr knows exactly who said what:
In remote meetings held through Microsoft Teams, Zoom, or Google Meet, speaker attribution is handled automatically. The accuracy of speaker attribution depends on the quality of metadata provided by the meeting platform. Some platforms, such as Zoom, may attribute all speech to a single participant or label segments as ‘Unknown Speaker’ unless additional platform features are enabled.
How Fostr Chooses an Attribution Method