Speech Strategy News

Speech Technology Group to make Microsoft speech technology more available

Server-based speech recognition and text-to-speech for telephone applications

Gerd Graumann, President, Speech Technology Group (STG), announced that his company can now resell Microsoft’s speech engine for telephone applications such as Interactive Voice Response (IVR) systems used in customer service. The Microsoft server-based speech software includes speech recognition and text-to-speech technology. Graumann said the Microsoft 64-Bit Speech Engine is the same basic technology used by Microsoft Tellme as a cloud-based technology.

Graumann said that a special agreement with Microsoft allows them to unbundle the speech engine from Microsoft Lync, Microsoft’s Unified Communications solution. Previously, he indicated, using Lync was the only way to get the engine. They have integrated the engine with MRCP (Media Resource Control Protocol, an industry-standard interface) and are now making the complete package available to the market.

The engine can be connected to any IVR or voice platform that supports MRCP (e.g., the Asterisk open-source PBX, Avaya, Genesys, some Cisco solutions). The engine supports standards SRGS (Speech Recognition Grammar Specification), SISR (Semantic Interpretation for Speech Recognition), and SSML (Speech Synthesis Markup Language), as well as MRCP.

Graumann said that STG can also help integrate the engine with solutions that don’t support MRCP. Languages used in IVR solutions, such as VoiceXML, can be used to drive applications on the platform.

STG suggests a list price of $210/port to end users for the complete speech recognition and TTS engine, all 26 languages (see table), support for large grammars, and the MRCP Connector. Graumann indicates that integrators who license the technology in volume will get a “hefty discount.”

MS ASR/TTS – 26 supported languages

Catalan Spain French
Chinese (Hong Kong) German
Chinese (Simplified) China Italian
Chinese (Traditional) Taiwan Japanese
Danish Korean
Dutch Netherlands Norwegian (Bokmal)
English Australia Polish
English Canada Portuguese Brazil
English India Portuguese Portugal
English UK Russian
English US Spanish Spain
Finnish Spanish Mexico
French Canada Swedish

STG has specialized in integrating speech technologies from outside vendors so that they work with multiple platforms and standards. They then support and resell those technologies. Graumann indicated that STG often brings partners together to offer an integrated solution that is larger than the sum of the parts. For example, one solution STG has supported is an application from Biddle Consulting Group called CritiCall, which uses TTS to simulate incoming 911 calls to train and test emergency dispatchers.

Speech Strategy News

Speech Technology Group to make Microsoft speech technology more available

Server-based speech recognition and text-to-speech for telephone applications

Gerd Graumann, President, Speech Technology Group (STG), announced that his company can now resell Microsoft’s speech engine for telephone applications such as Interactive Voice Response (IVR) systems used in customer service. The Microsoft server-based speech software includes speech recognition and text-to-speech technology. Graumann said the Microsoft 64-Bit Speech Engine is the same basic technology used by Microsoft Tellme as a cloud-based technology.

Graumann said that a special agreement with Microsoft allows them to unbundle the speech engine from Microsoft Lync, Microsoft’s Unified Communications solution. Previously, he indicated, using Lync was the only way to get the engine. They have integrated the engine with MRCP (Media Resource Control Protocol, an industry-standard interface) and are now making the complete package available to the market.

The engine can be connected to any IVR or voice platform that supports MRCP (e.g., the Asterisk open-source PBX, Avaya, Genesys, some Cisco solutions). The engine supports standards SRGS (Speech Recognition Grammar Specification), SISR (Semantic Interpretation for Speech Recognition), and SSML (Speech Synthesis Markup Language), as well as MRCP.

Graumann said that STG can also help integrate the engine with solutions that don’t support MRCP. Languages used in IVR solutions, such as VoiceXML, can be used to drive applications on the platform.

STG suggests a list price of $210/port to end users for the complete speech recognition and TTS engine, all 26 languages (see table), support for large grammars, and the MRCP Connector. Graumann indicates that integrators who license the technology in volume will get a “hefty discount.”

MS ASR/TTS – 26 supported languages

Catalan Spain French
Chinese (Hong Kong) German
Chinese (Simplified) China Italian
Chinese (Traditional) Taiwan Japanese
Danish Korean
Dutch Netherlands Norwegian (Bokmal)
English Australia Polish
English Canada Portuguese Brazil
English India Portuguese Portugal
English UK Russian
English US Spanish Spain
Finnish Spanish Mexico
French Canada Swedish

STG has specialized in integrating speech technologies from outside vendors so that they work with multiple platforms and standards. They then support and resell those technologies. Graumann indicated that STG often brings partners together to offer an integrated solution that is larger than the sum of the parts. For example, one solution STG has supported is an application from Biddle Consulting Group called CritiCall, which uses TTS to simulate incoming 911 calls to train and test emergency dispatchers.

css.php