Metreos Media Engine

The Metreos Media Engine is a software-only engine for native VoIP media processing that is seamlessly integrated as part of the Metreos VoIP application environment. It supplies ready-to-use, sophisticated VoIP media processing capabilities to all applications built using the Metreos 2400 — functions such as advanced IVR, conferencing capabilities, text-to-speech, speech recognition, and more (see below for more information).

By using the Metreos Media Engine to provide these capabilities, organizations can easily build much more useful, rich voice applications. In addition, use of the Metreos Media Engine offloads VoIP media processing from the IP PBX, saving precious slots and avoiding processing spikes that can impact PBX performance. The Metreos Media Engine supports a distributed architecture for clustering, scalable to meet the needs of the largest enterprise customers, and configurable to meet requirements for high availability.

Some of the key media processing capabilities offered by the Metreos Media Engine are:

IVR
The Metreos Media Engine supports advanced IVR functions such as playing prompts, recording audio, detecting and generating DTMF digits, and monitoring for silence and non-silence conditions. Additionally, developers can take advantage of complex termination and barge conditions when building applications to construct highly sophisticated interactive voice response systems.
Conferencing
Developers can leverage the native audio mixing capabilities of the Metreos Media Engine software when building applications. The Metreos Media Engine will perform active talker notification, n-way summing, and supports coach/pupil connections for applications that would like to perform a whisper or training function.
Text-to-Speech
Developers can now take advantage of integrated, easy-to-use text-to-speech media processing when building applications using the Metreos Visual Designer. The built-in TTS capability leverages a sophisticated on-board TTS engine and provides sophisticated voice generation capabilities. Developers can tune the TTS dictionary by adding their own pronunciations for words that should be pronounced differently. Finally, using the Speech Synthesis Markup Language (SSML) developers can control the pronunciation of phrases in a fine grained manner.
Speech Recognition
Developers can use voice recognition when building applications, for example, to recognize a spoken account code rather than requiring a user to enter the account code using a telephone keypad. The Metreos Media Engine natively supports the Nuance speech platforms.
Speaker Verification
Speaker verification can be used, for example, to authenticate a user based on his voice rather than requiring the user to authenticate by using a telephone keypad to enter a numeric account code and PIN. This biometric voiceprint authentication capability greatly extends a developer's options when building applications. Voiceprint authentication frees end-users from having to remember PIN codes and provides the developer with a secure, fault-tolerant method for authenticating end users.

Products
Products Product Overview VoIP Applications PhoneProxy ClusterMobility RapidRecord RemoteAgent Paging & Intercom Click-to-Talk VoIP Application Environment Application Runtime Media Engine Visual Designer Management Console VoIP Application Challenges Dial Tone Reliability Cost and Complexity Manageability How Metreos Is Different	Home » Products » Product Overview » VoIP Application Environment » Metreos Media Engine
	Metreos Media Engine The Metreos Media Engine is a software-only engine for native VoIP media processing that is seamlessly integrated as part of the Metreos VoIP application environment. It supplies ready-to-use, sophisticated VoIP media processing capabilities to all applications built using the Metreos 2400 — functions such as advanced IVR, conferencing capabilities, text-to-speech, speech recognition, and more (see below for more information). By using the Metreos Media Engine to provide these capabilities, organizations can easily build much more useful, rich voice applications. In addition, use of the Metreos Media Engine offloads VoIP media processing from the IP PBX, saving precious slots and avoiding processing spikes that can impact PBX performance. The Metreos Media Engine supports a distributed architecture for clustering, scalable to meet the needs of the largest enterprise customers, and configurable to meet requirements for high availability. Some of the key media processing capabilities offered by the Metreos Media Engine are: IVR The Metreos Media Engine supports advanced IVR functions such as playing prompts, recording audio, detecting and generating DTMF digits, and monitoring for silence and non-silence conditions. Additionally, developers can take advantage of complex termination and barge conditions when building applications to construct highly sophisticated interactive voice response systems. Conferencing Developers can leverage the native audio mixing capabilities of the Metreos Media Engine software when building applications. The Metreos Media Engine will perform active talker notification, n-way summing, and supports coach/pupil connections for applications that would like to perform a whisper or training function. Text-to-Speech Developers can now take advantage of integrated, easy-to-use text-to-speech media processing when building applications using the Metreos Visual Designer. The built-in TTS capability leverages a sophisticated on-board TTS engine and provides sophisticated voice generation capabilities. Developers can tune the TTS dictionary by adding their own pronunciations for words that should be pronounced differently. Finally, using the Speech Synthesis Markup Language (SSML) developers can control the pronunciation of phrases in a fine grained manner. Speech Recognition Developers can use voice recognition when building applications, for example, to recognize a spoken account code rather than requiring a user to enter the account code using a telephone keypad. The Metreos Media Engine natively supports the Nuance speech platforms. Speaker Verification Speaker verification can be used, for example, to authenticate a user based on his voice rather than requiring the user to authenticate by using a telephone keypad to enter a numeric account code and PIN. This biometric voiceprint authentication capability greatly extends a developer's options when building applications. Voiceprint authentication frees end-users from having to remember PIN codes and provides the developer with a secure, fault-tolerant method for authenticating end users.