- Contents
Interaction Designer Help
Record String Extended
This Telephony tool reads the contents of a string into an audio file (.WAV) using the Text-to-Speech (TTS) engine. This tool extends the options available over the Record String tool in that you can specify Voice Name, Speed, and other Optional Parameters.
For more information about installing, licensing, and configuring Interaction Text to Speech, see CIC Text To Speech Engines Technical Reference in the PureConnect Documentation Library.
Inputs
Text
The text that is recorded. This can also be the string variable.
Audio File Name (.WAV)
The path to and name of the .WAV file that is to be created. If you do not include any path information, the file will be generated in the directory specified by the "TTS Audio File Directory" server parameter. If no directory is specified, it defaults to whatever the Win32 API function GetTempPath uses, which typically returns "C:\TEMP" when called from a normal application and "C:\WINNT" when called from a service.
Voice Name
Name of voice to use to record the string.
Speed
This optional value is used to specify the approximate rate of speech. Since rates vary depending on what TTS engine is used, refer to the TTS provider documentation for more information.
Optional Parameters
To specify a name:value pair to control advanced MRCP properties, refer to Using MRCP Tools.
To use ITTS even if you have SAPI or MRCP as the default TTS provider, specify "I3TTS" as the optional parameter.
To specify a name:value pair to control advanced ITTS properties, refer to in Using ITTS Tools.
Outputs
Length of recording
An integer approximation of the playback length.
Exit Paths
Success
This step takes the Success exit path if the .WAV file is created.
Rejected
This step takes the Rejected exit path if one of the following occurs:
-
The server has no TTS engine installed.
-
The audio file already exists and is read-only.
-
The process does not have write permission in the directory specified for the audio file.
-
The string was empty or contained nothing that could be converted into "phonemes" (i.e. no audio was generated).