Measurable enhancements for DaaS & VDI end users and administrators

Proposed solution

Integrating the Fluendo Codec Pack into DaaS & VDI

The implemented solution involved integrating the Fluendo Codec Pack into the operating systems of various client endpoints, including Linux, Windows, IGEL OS, and Dell Wyse ThinOS. Crucially, the Fluendo Codec Pack included hardware acceleration plugins, leveraging the client device’s GPU horsepower to substantially reduce CPU load during audio and video content processing. This move guaranteed that client devices possessed optimized and fully licensed codecs for key industry formats, such as H.264 (including patent license compliance), MPEG-4, AAC, or MP3. Leveraging Fluendo’s specialized knowledge in GStreamer and multimedia experts, the codecs were specifically tuned for low-power thin client hardware, ensuring that the local decoding and rendering process was resource-efficient and fast.

By allowing the client to download and process the raw multimedia stream locally, the solution effectively bypassed the need for the VDI server to perform intensive video processing and recoding. This architectural refinement delivered multiple benefits: a reduction of load in the server (enabling greater user density), a better quality of video/audio for the end-user (achieving fluid HD/4K playback), and full legal compliance by managing the necessary patent licensing. The integration ensured interoperability with major VDI protocol vendors, including Citrix, Omnissa, Microsoft, and others.

Proposed solutions

Optimizing codec functionality for Chromium

Fluendo’s solution involved a dual-component strategy:

• Fluendo Codec Pack: A legally-compliant set of codecs providing the necessary H.264 and AAC support.
• FFmpeg Enabler for Chromium: A specialized component engineered to integrate the Codec Pack with the unique, internal version of FFmpeg used within the Chromium project, ensuring smooth interoperability.

This successful integration provided a robust, legally-compliant solution that allowed VDI users of major endpoint manufacturers (such as Dell, HP, among others) to seamlessly play web content requiring H.264 and AAC, thus solving the multimedia deficiency and enhancing the user experience.

Conceptual Design

Multimedia edge AI-powered virtual microphone with speech enhancement redirection for DaaS & VDI

Fluendo’s AI native-like voice enhancement toolset would operate as a virtual microphone on the user’s device, performing real-time noise suppression, voice enhancement, accent translation, and tone adjustment.

The enhanced audio stream would then be delivered into the DaaS & VDI session via standard microphone redirection (e.g., RDP/HDX/Blast), where any communication or meeting application would receive a clear, low-latency audio stream without requiring changes or additional setup.

For lower-performance or GPU-limited devices, a virtual recording component within the DaaS & VDI environment could provide the same processing capabilities server-side.

Leveraging Fluendo’s Multimedia Edge AI technology and Raven AI engine, this approach would emphasize Edge AI, real-time processing, and portability via GStreamer components to fit browser, Windows, Linux, macOS, iOS and Android endpoints.

Conceptual Design

DaaS & VDI GPU offloading via endpoint AI Superresolution optimized for thin clients

Fluendo’s AI-powered superresolution plugin would operate entirely on the client side to dynamically enhance video feeds, upscaling desktop content up to 4K resolution. The system would intelligently adapt to input video quality and integrate seamlessly with the DaaS protocol, native OS, browsers, and applications.

This approach would fundamentally decouple the end-user’s visual quality from network or server performance. It would remove the dependency on high-bandwidth connections and dramatically reduce the reliance on costly server-side GPU encoding, allowing businesses to deliver a premium experience over a standard internet connection.

Conceptual Design

Reference-free AI model for continuous user-perceived quality estimation

An AI model predicts reference-free video quality metrics (VMAF-like) directly from live desktop/video streams, enabling continuous QoS visibility and proactive anomaly detection.

The capability would be provided as a modular software library and a Fluendo AI GStreamer plugin using Raven AI Engine for inference, so that it could be embedded at the endpoint, broker, or server, and exported to existing observability stacks

The model would be trained to correlate strongly with human judgments and well-known objective metrics (such as VMAF), and validated across common DaaS workloads (text editors, spreadsheets, 3D tools, conferencing) to remain content-aware and low-overhead for real-time operation.

Conceptual Design

Dynamic, region-based enhancement

Our solution, built specifically for the remote desktop environment, will tap directly into the virtualized operating system’s window management APIs to understand the desktop composition in real-time. It will identify window boundaries, classify content type (text, graphics, video), and track motion vectors as they occur.

This rich contextual information will then be fed directly to our RDCT video codec and AI-based enhancements. This will allow them to bypass much of their own costly analysis and instantly apply the perfect quality and compression level to each region, dedicating maximum resources to critical areas while conserving them on static parts of the screen.