Your words are heard.
Your voice isn't.
Alterline enables real-time conversations where participants speak freely without exposing their real voice — built for journalists, investigators, and sensitive conversations.
Agent A: The documents confirm what we suspected
Built for the conversations that matter most
Every feature exists to protect the speaker without compromising the conversation.
Real-Time Voice Masking
Speech is transformed in real time so the original voice is never recorded or heard — only the altered voice is sent.
Identity Protection
No conversation logs are kept. Pair with Tor for full network-level anonymity. Join rooms without accounts or identity.
Voice Irreversibility
Even if a conversation is recorded, the original voice cannot be reverse-engineered. The transformation is a one-way function — your real voice is mathematically unrecoverable.
Natural Conversation
Low-latency audio keeps conversations fluid without relying on stored transcripts.
Who it's for
Alterline is purpose-built for high-stakes communication where anonymity isn't optional.
Journalists
Protect sources during recorded or live interviews.
Investigators
Conduct anonymous witness and informant calls.
Whistleblowers
Speak freely without fear of voice identification.
Ombuds
Receive confidential complaints while shielding the reporter’s identity.
Anonymous Groups
Participate in voice conversations without linking back to your real identity.
Privacy-Minded Individuals
Stay anonymous in any conversation — no voice fingerprint, no identity trail.
Threat Model
Alterline is designed around a specific set of threats. Here's what we protect against — and what falls outside scope.
What Alterline protects
- ✓Voice identification — Your original voice is never transmitted. Only the altered voice reaches other participants.
- ✓Recording & replay — Sessions are ephemeral. No voice data is stored on our servers after the call ends.
- ✓Participant correlation — No account or identity is required to join a room. Speakers are not linked across sessions.
- ✓Server-side exposure — Voice transformation happens on the relay engine before reaching any listener. The relay is self-hostable for full control.
What falls outside scope
- —Content of speech — Alterline masks your voice, not what you say. Be mindful of identifying details in your words.
- —Network metadata — IP addresses and connection timing may be visible to the server operator. Use a VPN or Tor for network-level anonymity.
- —Endpoint compromise — If a participant's device is compromised, local audio can be captured before transformation.
- —Speech-to-text provider — STT processes your spoken words and could be a point of exposure. The STT component is self-hostable for environments that require it.
- —Media transport (LiveKit) — LiveKit handles real-time audio routing and is open source, but a compromised instance could intercept streams. Self-host for full control of the media layer.
Simple, transparent pricing
Choose the plan that fits your needs. All plans include core privacy features. Upgrade or downgrade anytime.