Understand User Input Modes
Microsoft Unified Communications Managed API (UCMA) projects recognize input from users and create output in three modes:
Instant Messenger (IM)
DTMF (Dual-Tone Multiple-Frequency)
The mode supported at any time depends on the nature of the call, and which workflow activity is active. Design UCMA applications to support the expected call modes, which can be IM, speech or DTMF. To support IM calls, include activities that support IM, for example InstantMessagingQuestionAnswerActivity. To support speech and DTMF calls, include activities that support speech, for example SpeechQuestionAnswerActivity.
Another way of understanding application mode is to consider whether the application makes calls or receives calls. Communications workflow projects support both modes for speech and instant messaging.
Communications workflow applications can recognize speech, and emit speech as prompts. To support speech mode, add one or more speech grammars and the appropriate speech activities.
To create new speech recognition grammars in UCMA applications, use the Grammar class to create Speech Recognition Grammar Specification (SRGS) grammars, as described in Authoring Grammars.
To emit verbal prompts in a communications workflow application, use either of these activities:
Communications workflow applications can recognize text input, and emit text as prompts.
Instant Message Text Recognition
Speech grammars perform input recognition in instant message (IM) mode. To support IM mode, add one or more speech grammars and the appropriate IM activities:
To emit text prompts in a communications workflow application, use either of these activities:
A workflow activity that recognizes speech input will also recognize DTMF input if a DTMF grammar is added to the grammar collection. To create new DTMF recognition grammars in UCMA applications, use the Grammar class to create SRGS grammars, as described in Authoring Grammars.