Confirm / Resources / Behind the build

Why Confirm selected Pindrop for voice authentication

Choosing the engine behind member identity is not a decision you make on a feature list. Here is how we evaluated the field, and why we landed on Pindrop.

When we set out to add voice authentication to Confirm, we treated the vendor choice as the most consequential one in the roadmap. The engine would sit on the critical path of every member call. It had to be accurate, fast, and durable against threats that did not exist a year ago.

What we were testing for

  • Accuracy at the edges — not just clean calls, but noisy lobbies, poor cell connections, and short utterances.
  • Deepfake and replay resistance — the ability to tell a live human from a synthetic or recorded voice, in real time.
  • Scale and track record — a platform already trusted by large financial institutions, not a pilot-stage product.

What set Pindrop apart

Pindrop's liveness and deepfake detection — Pindrop Pulse — was the clearest differentiator. It scores each call for signs of synthetic, replayed, or machine-generated audio, which is exactly the threat that is growing fastest against call centers. Combined with mature voiceprint matching and a deployment history across major banks, it cleared the bar on every axis we cared about.

Right-sized for the institutions we serve

Enterprise-grade technology is only useful to a community institution if it can actually be deployed and paid for. Part of our job was packaging Pindrop's capabilities so a credit union under $3 billion in assets gets the same protection a national bank does — without an enterprise integration project.

Curious how the underlying detection works on your call traffic? Ask us for a walkthrough of Pindrop Pulse scoring inside Confirm.

See Confirm running in your core.

Tell us about your institution and we’ll get back to you within 24 hours.

Request more info