NSC Releases Version 3.8 of its Hardware Based Speech Recognition Engine; Version 3.8 introduces new product features and support of market standards.

RISHON LEZION, Israel -- NSC - Natural Speech Communication Ltd., developer of hardware based speech recognition products for the telephony market, has announced the release of Version 3.8 of its hardware based speech recognition engine.

The new version, an ASR Server running on NSC's family of PCI boards (NSCBoard(TM)), introduces new product features and improved support for market standards and customer requirements.

NSC's Hardware based ASR product

The NSCBoard(TM) is a family of boards that includes 2, 4, 12 or 30 DSPs. The NSCBoard(TM)-30 is a high density, cost-effective, DSP based board, designed to deliver up to 120 concurrent ASR channels in a single PCI slot. Each DSP is autonomous and can be independently configured for a different language; for each ASR channel request, a different vocabulary and grammar can be used. Several ASR channels can run on a single DSP, depending on the speech recognition task complexity. NSC has two new hardware developments:

--NSC has introduced a new board to its family of boards: NCSBoard2(TM) - to support small to medium business with up to 8 ASR channels on a single board.

--The NSCBoard(TM) family of boards was upgraded to support the PCIU standard.

ASR Functionality

--The new version supports the ability of the engine to return multiple recognition results (N Best) to the application with a confidence level for each returned result.

--Included in NSC's speech recognition engine, echo cancellation was implemented as part of the ASR resource. Echo cancellation is a needed feature to support "barge-in" capabilities - the ability to speak during the application voice prompts. The speech recognition engine includes an echo-cancellation capability, thus eliminating the need for having this feature on the telephony board.

--Languages: Version 3.8 supports speech recognition in the following languages: North American English, UK English, Russian, German, Hebrew and Arabic.

--Standards: The functionality of NSC's Server and APIs were enhanced to fully support VXML and MRCP specifications.

Key Word Spotting Engine

A Key Word Spotting (KWS) engine is included in Version 3.8 with the capability to spot keywords spoken while monitoring on-going calls in real time or recorded calls.

Unlike speech recognition for telephony speech-driven services, the KWS engine is active for the whole call duration, looking for specific words within a long spontaneous and natural conversation between two or more speakers.

Using NSC's high density family of speech recognition engines, the NSCBoard(TM), the new KWS engine can provide real-time monitoring capability of thousands of simultaneous calls with simple system architecture.

Improved Grammar functionality

--Version 3.8 brings Built-In grammars according to the VXML specifications. These are off-the-shelf grammars of standard and common phrases such as times, dates and natural numbers.

--Version 3.8 now fully supports ABNF grammar format.

--The new version also introduces support to Stochastic Grammars.

--Activation/de-activation of grammars is possible.

Server and APIs

The NSC server is capable of handling any combination of board types and sizes while the same speech recognition engine runs on all boards. Version 3.8 introduces more advanced features such as:

--A flexible configuration tool allowing autonomous definition of the NSCBoard(TM) speech recognition resources.

--New API commands that enable more flexible management of the NSC servers by the application. This means that the application can determine, in run time, whether to launch additional servers or shut-down running servers, according to the need.

--Can be operated on LINUX red-hat operating system. Overall - the NSC server supports Windows NT, Windows 2000, Windows 2003, Windows XP and LINUX.

--Higher Flexibility in defining recognition parameters. For example, the duration time of utterance can now support up to 30 seconds.

--The ability for a run time configuration of the resources (remote/local) of the feeds to be recognized. The source feed for the NSC engine can be provided from a local host through physical direct connection to the telephony board or from a remote source via Ethernet.

"Our new version with its additional functionality and flexibility is another step in our way to offer the market a simple but strong, stand alone speech recognition product which can be easily integrated and configured according to the changing needs of the application developers in order to enable a widespread application of speech recognition products," said Dr. Ami Moyal, CEO of NSC.

About NSC

NSC specializes in speech recognition technology for telephony systems. NSC's solution is based on hardware - PCI boards that are designed around dedicated processors (DSP). This unique approach by NSC allows very high density at low cost with simple system architecture. NSC's hardware based solutions performs speech recognition without any need for CPU resources, with minimal footprints, side by side with maximum control for the users in optimizing system resources for their needs. The company is active in the US, Europe and Middle East and involved in the implementation of speech recognition and key-word spotting based applications through business partners: IVR and VXML platforms, telephony application developers, system integrators, recording platforms developers and customers such as cellular networks operators.

http://www.nscspeech.com