ringback — ringback-alert / ringback-voice MCP servers
Copyright (c) 2026 the project contributors

This product is licensed under the Apache License, Version 2.0 (the "License");
see the LICENSE file. You may not use this project except in compliance with
the License.

================================================================================
IMPORTANT LICENSING CAVEAT — the voice feature depends on GPL software
================================================================================

The phone-VOICE component (voice_agent.py, voice_mcp.py) imports **pjsua2**, the
Python bindings of **PJPROJECT / PJSIP**, which is licensed under
**GPL-2.0-or-later OR a commercial license** from Teluu Ltd.

Because of this, if you DISTRIBUTE a combined work that links/imports pjsua2,
that combined work is subject to GPL obligations (or you must obtain a
commercial pjproject license). The Apache-2.0 license on THIS repository's own
source does not, and cannot, relax pjproject's GPL terms.

Practical guidance:
  * Using the voice feature privately for yourself: no problem.
  * Redistributing a bundle that includes/links pjsua2: comply with the GPL
    (share source under GPL-compatible terms) OR buy a pjproject commercial
    license. See https://github.com/pjsip/pjproject and https://www.pjsip.org.
  * The phone-ALERT component (server.py: ntfy / Pushover / baresip) does NOT
    depend on pjproject and is unaffected by this caveat.

This is informational, not legal advice. If you plan to redistribute, consult
the upstream licenses and, if needed, a lawyer.

================================================================================
THIRD-PARTY COMPONENTS
================================================================================

Runtime dependencies (installed by setup.sh / setup-linux.sh / Dockerfile; NOT bundled
in this repo):

  PJPROJECT / pjsua2  — GPL-2.0-or-later OR commercial — https://www.pjsip.org
      Used by ringback-voice for the SIP call + media engine.

  whisper.cpp         — MIT License — https://github.com/ggerganov/whisper.cpp
      Used for on-device speech-to-text. Model files (ggml-*.bin) are
      downloaded from Hugging Face and are not redistributed here.

  baresip             — BSD-3-Clause — https://github.com/baresip/baresip
      Used by ringback-alert for the SIP ring + chat message.

  ntfy                — Apache-2.0 — https://github.com/binwiederhier/ntfy
      Optional self-hosted push server.

  Piper (piper-tts)   — MIT License — https://github.com/rhasspy/piper
      Default cross-platform neural text-to-speech. Voice models (.onnx) are
      downloaded from Hugging Face (rhasspy/piper-voices) and not redistributed here.
  macOS `say`         — Apple system text-to-speech (part of macOS; the macOS fallback).
  espeak-ng           — GPL-3.0-or-later — https://github.com/espeak-ng/espeak-ng
      Optional fallback TTS engine on Linux/Windows.
  ffmpeg              — LGPL/GPL depending on build — https://ffmpeg.org
      Used to transcode TTS output to the call's audio format.

  Python: mcp (MIT), httpx (BSD-3-Clause), OpenSSL (Apache-2.0).

Each third-party component is the property of its respective owners and is
governed by its own license. Refer to each project for full terms.