Voice codec capability can triple speech transcoding density

The high performance Intel processors with integrated graphics, combined with software-based speech processing, can yield up to 3,000 speech encoding sessions per processor, bringing new efficiencies to voice transcoding applications in operator networks and enterprises.

Stephen Dow, President, Artesyn Embedded Technologies, said: “This integrated solution will deliver tremendous flexibility for our customers, who need to deploy virtual network functions or VNFs across multiple hardware platforms. Our approach scales from edge to core with the highest density per RU to help carriers lower their OpEx.”

A proof-of-concept demonstration on the Intel booth (8.1E41) will enable processing of up to 140,000 channels of G.711 to G.729 transcoding in an Artesyn carrier-grade server, the highest density per rack unit in a standards-based server. This solution combines software expertise from Vantrix and Artesyn’s high density computing hardware system, which leverages removable SharpStreamer PCI Express media accelerator cards that can be used in a range of Artesyn or third party platforms.

“By using Intel processors, Artesyn and Vantrix have the performance needed to deliver impressive speech transcoding capabilities," said Sandra Rivera, Vice President and General Manager, Network Platforms Group, Intel. "This gives customers the flexibility to deploy applications such as VoLTE, session border controllers, and media gateways as virtual network functions.”

“Vantrix is excited to showcase next-generation real-time speech capabilities on Intel processors. These ultra-high-density processing improvements available from Artesyn, together with our cloud-based media processing solution, will help operators and enterprises be more profitable and efficient for real-time communications, including via WebRTC and VoLTE," added Jean Mayrand, Chief Executive Officer, Vantrix.