User Language Support for English (United States)
This content is no longer actively maintained. It is provided as is, for anyone who may still be using these technologies, with no warranties or claims of accuracy with regard to the most recent product version or service release.
User languages supported by Speech Server are identified by their culture names, as defined in the System.Globalization.CultureInfo object that represents the language. The culture name for the English language in the locale of the United States is en-US and its display name is English (United States).
The System.Globalization structure also provides a culture identifier, called an LCID, for each language that is supported by Windows. But Speech Server supports certain languages that are not supported by all versions of Windows and these languages cannot be accessed by their LCID. For this reason, it is recommended that application authors reference user languages by their culture name rather than by LCID.
Support for user interaction in English (United States) consists of the following elements.
Voice Mode Grammar Library
The Voice Mode Grammar Library for English (United States) is a collection of ready-to-use rules and rulesets designed to recognize commonly used types of user voice input such as dates, times, currency units, numbers, and confirmatory responses. Use the rules as-is for recognizing simple input or build more complex grammars that reference these rules. Use the library for all types of Web-based and managed code voice response applications in Speech Server.
When an application uses the library, all the rules in the library are loaded into the recognition engine as inactive. Recognition output for a rule is not returned unless the application activates the rule prior to using it. A specific rule is activated by adding a ruleref element that references the rule.
The grammar library is contained in an XML-formatted file named Library.En-us.grxml at the following location:
%ProgramFiles%\Microsoft Office Communications Server 2007 Speech Server\SDK\Templates\1033\
When you write new grammars for your application, you can add ruleref elements that reference rules in the grammar library or copy rules from the library and paste them into your grammars.
DTMF Mode Grammar Library
The DTMF Mode Grammar Library for English (United States) is a collection of ready-to-use rules designed to recognize numeric values associated with the DTMF pitches that are generated by a user pressing the keys of a telephone keypad. Use this library similarly to the way in which you use the Voice Mode Grammar Library.
The DTMF library is contained in an XML-formatted file named DTMFRules.grxml at the following location:
%ProgramFiles%\Microsoft Office Communications Server 2007 Speech Server\SDK\Templates\1033\
Remarks
- Modifying the grammar library file is not recommended. If modification is necessary, copy the library file to a new file (for example, MyRules.grxml) and modify the new file.
- Besides the grammars that developers create, SALT voice response applications that use ASP.NET Application Speech Controls also use additional voice mode and DTMF mode grammar libraries. These libraries are saved in a compiled .cfg binary format and are located in the following directory:
%SystemDrive%\Inetpub\wwwroot\aspnet_speech\v2.0.3400.0\client_script\1033\
These libraries are not available to developers for direct use in application development. However, it is important to know that Speech Server loads and caches these library files the first time an application containing Application Speech Controls references them.
Phoneme Table
Custom pronunciations for text-to-speech (TTS) engines are expressed in phonemes specific to the language of the engine. Click the following link for the phoneme table for English (United States).