TMCnet - World's Largest Communications and Technology Community



Speech Recognition and Text to Speech Provider Loquendo Powers MRCP Server


TMCnews Featured Article

July 26, 2011

Speech Recognition and Text to Speech Provider Loquendo Powers MRCP Server

By Jamie Epstein, TMCnet Web Editor

Loquendo (News - Alert), a provider of speech technologies including speech recognition and text to speech software, has just released an updated version of the MRCP Server (LMS 7.2). The newly launched version is based upon the foundation of the original infrastructure but now features multiple added capabilities that can help to dramatically increase scalability. This new highly efficient and fully scalable architecture improves the supported port density, compared to previous versions the LSS 7.0 and the LMS 7.1, a press release detailed.

The new architecture, which boasts the ability to perform multiple functions simultaneously, has an internal controller that can make sure the correct amount of internal workload is distributed evenly, thus maximizing port density. The new server is available for both Windows and Linux.

Thanks to this new server, memory requirements are greatly reduced and, due to its multi-process architecture, physical memory above the 4GB limit can be easily digested by 32-bit architectures, elevating the level of scalability when deploying next generation multi-core processors.

In addition, the release stated that a new wizard tool is also incorporated into the server, which enables the user to easily configure the LMS 7.2; as it also automatically configures the appropriate number of “MRCP server Core” process instances.

This new version of Loquendo MRCP Server can support the latest versions of Loquendo ASR (7.10), as well as Loquendo TTS (7.20), further encouraging users to see the advantages of the multitude of new features offered by those two new releases.

Along with 8 kHz a-law and µ-law audio formats, LMS 7.2 now also supports the 16 kHz linear encoding. The release revealed that a new recognition object (RO) caching feature enables ROs to be simply shared between different MRCP sessions, so that every grammar provided will be compiled only once. Also, ROs can now be pre-compiled and loaded permanently.

Loquendo MRCP (Media Resource Control Protocol) Server is perfect for creating large-scale telephony applications through use of Loquendo multilingual speech, and can be configured as TTS-only, ASR-only, or both TTS and ASR—whatever the user desires.

In conclusion, Loquendo MRCP Server can be seamlessly integrated with all major speech platforms and can help any organization increase scalability, while reducing complexities.

Jamie Epstein is a TMCnet Web Editor. Previously she interned at News 12 Long Island as a reporter's assistant. After working as an administrative assistant for a year, she joined TMC (News - Alert) as a Web editor for TMCnet. Jamie grew up on the North Shore of Long Island and holds a bachelor's degree in mass communication with a concentration in broadcasting from Five Towns College. To read more of her articles, please visit her columnist page.

Edited by Rich Steeves

Technology Marketing Corporation

2 Trap Falls Road Suite 106, Shelton, CT 06484 USA
Ph: +1-203-852-6800, 800-243-6002

General comments: [email protected].
Comments about this site: [email protected].


© 2022 Technology Marketing Corporation. All rights reserved | Privacy Policy