video1.0<iframe src="https://www.loom.com/embed/0417ec8d70df454d9e68e618ee5a202b" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/0417ec8d70df454d9e68e618ee5a202b-6854f605caf8b130.gif141.845333This Loom demonstrates a voice authentication app that verifies a user’s voice and authorizes an agent. It sets up two baseline voice recordings, extracts speech to text, stores it in a vector embedding database, and matches it against the recording and text to validate access. The Guild AI agent then runs in the background to create a file in the Documents folder named meetingfortomorrow.txt when verification succeeds. The speaker adds that voice encoding includes different pitches such as low and middle (and optionally high) and that the encoded voice serves as the source of truth for validation.