After almost a year of development, Xiph.Org, an organization dedicated to the development of free video and audio codecs, has unveiled the release of the Opus 1.5 audio codec, providing high-quality encoding and minimal latency for both high-bitrate streaming audio compression and voice compression in bandwidth-constrained applications VoIP telephony. The encoder and decoder reference implementations are licensed under the BSD license. The complete specifications for the Opus format are publicly available, free, and approved as an Internet standard (RFC 6716).
The codec is created by combining the best technologies from Xiph.org's CELT codec and Skype's open source SILK codec. In addition to Skype and Xiph.Org, companies such as Mozilla, Octasic, Broadcom and Google also took part in the development of Opus. The patents involved in Opus are granted by the companies involved in the development for unlimited use without payment of royalties. All intellectual property rights and patent licenses related to Opus are automatically delegated to applications and products using Opus, without the need for additional approval. There are no restrictions on the scope and creation of alternative third-party implementations. However, all rights granted are revoked in the event of patent proceedings affecting Opus technologies against any user of Opus.
Opus features high coding quality and minimal latency for both high-bitrate streaming audio compression and voice compression for bandwidth-constrained VoIP telephony applications. Previously, Opus was voted the best codec at 64Kbit (Opus outperformed competitors like Apple HE-AAC, Nero HE-AAC, Vorbis and AAC LC). Products that support Opus out of the box include the Firefox browser, the GStreamer framework, and the FFmpeg package.
Main features of Opus:
- Bitrate from 5 to 510 Kbit/s;
- Sampling frequency from 8 to 48KHz;
- Frame duration from 2.5 to 120 milliseconds;
- Support for constant (CBR) and variable (VBR) bitrates;
- Support for narrowband and wideband audio;
- Voice and music support;
- Stereo and mono support;
- Support for dynamic setting of bitrate, bandwidth and frame size;
- Ability to restore the audio stream in case of frame loss (PLC);
- Support up to 255 channels (multi-stream frames)
- Availability of implementations using floating and fixed point arithmetic.
Key innovations in Opus 1.5:
- The encoder and decoder have expanded the use of machine learning technologies.
- Implemented a Deep Redundancy (DRED) mechanism that uses machine learning to recover audio fragments lost due to packet loss.
- Improved restoration of the audio stream in case of frame loss using the Deep PLC (Packet Loss Concealment) machine learning engine.
- Improved voice quality at low bitrates (~ 6 Kbps).
- Added additional optimizations using the AVX2 (x86) and Neon (ARM) instruction sets.
- Added support for the 4th and 5th order Ambisonics surround sound format, which takes into account sound propagation not only in the horizontal plane, but also in the vertical (to determine the sound source from above or below).
Hot on its heels, Opus 1.5.1 was released, which fixed problems with building using the Meson build system (building using CMake was not affected by the problem).
Source: opennet.ru