Dmytro Bogovych
c8d9bad845
The previous src/libs/g722 was spandsp's G.722 offered only under GPL-2 or LGPL-2.1. That blocks shipping rtphone statically linked in a closed binary (the freeware VQ for Asterisk engine) without a relink offer. Asterisk's codecs/g722 is the same code: Steve Underwood placed his contributions in the public domain there, and the CMU 1993 code under it is "completely unrestricted" for any use. src/libs/g722/README.md records the provenance and both statements. It also fixes the decoded level. Asterisk's copy halves the encoder input and doubles the decoder output, so streams from Asterisk and ffmpeg decode at their original level. The old copy decoded them 6 dB too quiet (ITU-convention streams from pjmedia 12 dB), and its encoder clipped input above -6 dBFS. Local changes to Asterisk's files: - the decoder saturates its output instead of casting. Asterisk's cast wraps: 259 samples in 11 s of speech peaking at -0.2 dBFS came out with the opposite sign; - g722.h includes <stdint.h> and maps __inline__ for MSVC (the removed g722_inttypes.h used to). Dropped: g722_bitstream.*, g722codec.c and three helper headers, none used. The API is unchanged (g722_encode_init / g722_decode, same options). Measured in vq-core (16 simultaneous replayed calls, NISQA speech): - G.711 control calls score identically; - normal-level G.722: SilentCall -0.09..-0.14, DeadAir-01 -0.06..-0.09, MOS unchanged; - loud G.722 with a clipped source now shows AmpClipping 0.02-0.03 and MOS 3.89/4.28 instead of 4.45: the codec overshoots on the clipped peaks at the correct level, which the old -6 dB hid; - results are identical across runs (the old decoder varied by up to 0.30 MOS). G.722 still loses about 44% of its packets in AudioReceiver because G722Codec::info() reports an 8 kHz sample rate for 16 kHz audio. That is fixed separately. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
RTPhone Platform
RTPhone is a comprehensive real-time communication (RTC) platform that provides a complete software stack for building VoIP/SIP-based communication applications. Developed by VoIP Objects (Sevana), RTPhone delivers production-ready voice communication capabilities with extensive codec support and cross-platform compatibility.
Overview
RTPhone serves as a static library (librtphone.a) that can be integrated into larger telephony and communication systems. It provides a JSON-based command interface for easy integration and control, making it suitable for building softphones, PBX systems, WebRTC gateways, and carrier-grade voice solutions.
Key Features
Audio Codec Support
RTPhone supports an extensive range of audio codecs:
Standard Codecs:
- G.711 (A-law/¼-law)
- G.722 (16kHz wideband)
- G.729
- GSM (Full Rate, Half Rate, Enhanced Full Rate)
- iLBC (20ms/30ms)
- ISAC (16kHz/32kHz)
Advanced Codecs:
- AMR-NB/AMR-WB (Adaptive Multi-Rate Narrowband/Wideband) - please be aware - there is no patents for AMR codecs usage included ! You should acquire them on your own.
- EVS (Enhanced Voice Services) - 3GPP's latest codec. Again - please be aware - there is no patents for EVS codec usage included ! You should acquire them on your own.
- Opus - Modern low-latency codec
- Speex (with acoustic echo cancellation)
Codec Features:
- Bandwidth-efficient and octet-aligned modes
- IuUP (Iu User Plane) protocol support for 3G networks
- Dynamic codec switching
- Packet loss concealment (PLC)
- Comfort noise generation (CNG)
Network & Protocol Support
SIP Features:
- Full SIP 2.0 implementation via reSIProcate
- Multiple transport protocols (UDP, TCP, TLS)
- Registration, authentication, and session management
- SIP MESSAGE, presence, and REFER support
Media Transport:
- RTP/RTCP for media streaming
- SRTP for secure media
- ICE for NAT traversal with STUN/TURN support
- WebRTC integration components
- IPv4 and IPv6 support
Cross-Platform Audio Support
- DirectSound/WMME (Windows)
- Core Audio (macOS/iOS)
- ALSA/PulseAudio (Linux)
- Oboe (Android) for low-latency audio
- PortAudio fallback support
Audio Quality Features
- 48kHz sample rate support
- Acoustic Echo Cancellation (AEC)
- Audio resampling and format conversion
- Multi-channel audio mixing
- Perceptual Voice Quality Assessment (PVQA)
Architecture
The platform is organized into several core modules:
- Engine/Agent: JSON-based command interface
- Engine/Endpoint: SIP user agent implementation
- Engine/Media: Audio codec management and processing
- Engine/Audio: Cross-platform audio I/O handling
- Engine/Helper: Utility functions (networking, logging, threading)
Supported Platforms
- Linux (x64, ARM/Raspberry Pi)
- Windows (32/64-bit)
- macOS
- Android (with Oboe integration)
- iOS
Building
RTPhone uses a CMake-based build system with cross-compilation support:
Linux
python3 build_linux.py
Android
python3 build_android.py
# or
./build_android.sh
Dependencies
- CMake 3.10+
- OpenSSL 1.1+
- Boost libraries
- Platform-specific audio libraries
Recent Updates
Recent development has focused on:
- AMR codec parsing and decoding improvements
- Octet-aligned mode fixes for AMR-WB
- RTP SSRC handling enhancements
- Build system optimizations
- Code modernization to C++20
Use Cases
RTPhone is good for building:
- VoIP softphones and mobile applications
- PBX and telephony server systems
- RTP proxies
- Carrier-grade voice communication platforms
- 3GPP/IMS-compliant systems
Security
RTPhone includes comprehensive security features:
- OpenSSL 1.1 integration for encryption
- TLS transport layer security
- SRTP media encryption
- Certificate management support
Integration
The platform provides a developer-friendly interface with:
- Event-driven architecture
- Comprehensive logging system
- Modern C++20 codebase
For detailed integration instructions and API documentation, please refer to the source code and header files in the /src/engine/ directory.
License
Our source code is licensed under the MPL license. Naturally, any third-party components we use are subject to their respective licenses.