AppsAI Agent Firewall

Voice and caller security

Voice watermark

When the firewall is on, every Autessa agent speaks with an inaudible watermark, so if a caller replays a recording of your agent, for example to pass a voice check or replay a code it read out, the firewall recognizes the watermark and treats the audio as synthetic.

Turn on the firewall
voice/score · uploaded clipreplay checkSAMPLE
score0.91
labelSYNTHETIC
watermarktrue · Autessa agent voice

A caller replayed a recording of your own agent reading out a code. The firewall treated it as synthetic.

The problem

Voice agents read things out loud. A recording of your own agent is perfect replay material for an attacker.

How voice watermark works

STEP 1

Agents speak with a mark

The watermark doesn’t change how the agent sounds.

STEP 2

Replays are recognized

Audio carrying the mark is treated as synthetic.

STEP 3

Check any clip

The voice score API reports whether a recording carries the watermark, so you can confirm a clip came from an Autessa agent.

Every setting

Voice watermark

On by default; turn off on the Settings page.

Who can do what

Admins

Turn the watermark on or off.

What to know

  • It’s most reliable in voice chat and on recordings. Heavy phone compression can weaken it, so treat a missing watermark on a call as unknown, not proof of a live voice.

More in Voice and caller security

Voice deepfake detection →Every thing a caller says on a phone call or in voice chat gets a synthetic-voice score from 0 to 1, labeled Real, Synthetic or Unsure, and a policy action applies at 0.5 or above.Caller ID check (STIR/SHAKEN) →On every inbound call the firewall reads the carrier’s STIR/SHAKEN result and applies your action for Verified, Partly verified, Not checked and Failed numbers.All of Voice and caller security →

FAQ

Voice watermark: questions

What is AI voice watermarking?

An inaudible signal embedded in generated speech that a detector can recognize later. Autessa marks its agents’ voices so replayed agent audio is caught.

Turn on voice watermark.

Autessa agent voices carry an inaudible watermark, so replayed agent audio is treated as synthetic, and the voice score API reports whether a clip came from an Autessa agent. Start in Monitor and see what it would catch on your real traffic.

Turn on the firewall

Starts in Monitor · nothing changes for your customers until you enforce

100% customizable, code and all

Make Autessa AI Agent Firewall fully your own.

Use it out of the box, then change anything. Ask Autobot, Autessa’s AI builder, for a new field, screen, report, workflow or agent and it builds it into your copy of the app. Or open the code and edit it yourself. Nothing is locked: the app, its data and its code are yours.

Start free and make it yours