The release of PipeWire 0.3.33 has been published, developing a next-generation multimedia server that replaces PulseAudio. PipeWire expands the capabilities of PulseAudio by providing tools for working with video streams, low-latency audio processing, and a new security model for managing access at the individual device and stream level. The project is supported in GNOME and is already used by default in Fedora Linux. The project code is written in C and is distributed under the LGPLv2.1 license.
Key changes in PipeWire 0.3.33:
- Automatic switching between Bluetooth profiles HSP (Headset Profile) and A2DP (Advanced Audio Distribution Profile) has been implemented.
- In the Pro Audio profile, support for virtual audio sources and output devices has been enhanced.
- Improved format modifier agreement when multiple clients share Direct Memory Access (DMA-BUF) buffers.
- Support for simultaneous use of multiple sampling rates has been implemented in the multimedia nodes section (disabled by default).
- In the compatibility layer with PulseAudio, the module module-switch-on-connect has been implemented to emulate PulseAudio behavior when new devices are connected.
Remember, PipeWire expands the use of PulseAudio by processing any multimedia streams and is capable of mixing and redirecting video streams. PipeWire also provides capabilities for managing video sources, such as video capture devices, webcams, or content being output by applications. For example, PipeWire allows several applications to collaborate on a webcam and addresses issues with secure screen content capture and remote access to the screen in a Wayland environment.
PipeWire can also serve as an audio server, providing minimal latency and offering functionality that combines the capabilities of PulseAudio and JACK, including addressing the needs of professional audio processing systems that PulseAudio could not cater to. Additionally, PipeWire provides an advanced security model that allows management of access at the level of individual devices and specific streams, simplifying the organization of audio and video passthrough from and to isolated containers. One of the main goals is to support self-sufficient applications in Flatpak format and to operate within the graphical stack based on Wayland.
Main Features:
- Capture and playback of sound and video with minimal latency;
- Tools for real-time audio and video processing;
- Multiprocess architecture allowing collaborative access to content across multiple applications;
- A model for processing based on a multimedia node graph with support for feedback loops and atomic graph updates. It allows for handlers to be connected both inside the server and by external plugins;
- An efficient interface for accessing video streams via file descriptor passing and accessing audio through shared ring buffers;
- Ability to process multimedia data from any processes;
- A GStreamer plugin to simplify integration with existing applications;
- Support for isolated environments and Flatpak;
- Support for plugins in SPA (Simple Plugin API) format and the ability to create plugins that operate in hard real-time mode;
- A flexible system for negotiating media formats and allocating buffers;
- Using a single background process for routing audio and video. Ability to operate as an audio server, a hub for providing video to applications (e.g., for the gnome-shell screencast API), and a server for managing access to video capture hardware.
Source: opennet.ru
