Speech Output References
This content is no longer actively maintained.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.
Speech Server supports two types of spoken output: text-to-speech (TTS) synthesis and recorded prompts.
Speech Server uses TTS as the default type of speech output. The main advantage of TTS is the convenience for the application author, and its main disadvantage is its unnatural sound. Speech Server also supports a method of speech output that uses recorded prompts. The advantage of recorded prompts is their natural sound, and their disadvantage is the effort expended recording the prompts.
SALT voice response application authors can specify prompt text in the SALT prompt Element, in the InlinePrompt and PromptSelectFunction properties of the DialogPrompt Speech Control, and in the QuestionPrompt and PromptSelectFunction properties of Application Speech Controls. The prompt text can also include a SALT value Element, which gets text values from HTML elements on the page, and a SALT content Element, which can specify prerecorded sound files and other types of content.
Managed code voice response application authors can specify prompt text or Speech Synthesis Markup Language (SSML) content using a PromptBuilder object.
When a TTS engine pronounces words or acronyms improperly, authors can use SSML markup to specify proper pronunciations. The pronunciations are expressed in a set of phonemes specific to the language of the TTS engine. Phoneme tables are available for the following installed user languages.
|Phoneme Table for Spanish (United States)||SpeechServer2007 Collection|
|Phoneme Table for German (Germany)||SpeechServer2007 Collection|
|Phoneme Table for French (Canada)||SpeechServer2007 Collection|
|Phoneme Table for English (United Kingdom)||SpeechServer2007 Collection|
|Phoneme Table for English (United States)||SpeechServer2007 Collection|