Radisys Unveils New Advanced Speech Recognition Support for Enhanced In-Call Speech Services
Radisys Corporation, a provider of open telecom solutions, announced new features for its MediaEngine virtualized media server to enable advanced in-call services, allowing Communication Service Providers (CSPs) to implement speech recognition interfaces at dramatically lower costs than previous solutions. These new features allow CSPs to accelerate new speech-driven service introductions by utilizing off-the-shelf Intel Architecture servers, advanced media processing with MediaEngine, and a combination of advanced speech processing engines, to significantly lower OpEx and CapEx, all within a well-defined open API framework.
Radisys’ MediaEngine offers new features that enable CSPs to overcome the challenges associated with the traditional in-call speech-enabled applications, and deploy new solutions, including enhanced customer service or application navigation (like in-conference commands for more hands-free capability), or to develop new revenue streams by using speech analytics to build actionable customer engagements. These new features include:
- The keyword detection embedded in the media server delivers a small footprint keyword or key phrase speech recognition engine, allowing for an “always on” solution that reduces costs of deploying speech-enabled services by over 90%.
- Integrated support for multi-vendor best-in-class server based in-network or cloud speech platforms for advanced speech interactions including natural language conversational recognition, and text-to-speech processing, allowing service providers to pick the best technology for their particular use case.
- Voice Quality Enhancements (VQE) overcome telecom network or environmental conditions that can reduce in-call recognition accuracy.
- Radisys’ MediaEngine software with embedded real-time advanced speech processing and media analytics allows CSPs to create, deliver, and monetize new applications by delivering cost-effective advanced technologies that were previously cost prohibitive for many innovative applications.
Google announced the general availability of cloud speech-to-text and multi-channel recognition. Additionally, new languages are available and discounts are available.