STVoiceFlow framework is a powerful framework providing STVoiceFlowRunner and STMediaRunner modules for processing and executing conversational Voiceflows with support of powerful media serivces. These modules provide SDKs powered with real-time event callback engines that allow programs to execute complex conversational speech interactions with their users.

STVoiceFlow framework features:

  • Interpretation and processing of Voiceflows comprised of Voiceflow Modules and Audio Prompt Modules.
  • Adaptation of Voiceflow processing behaviour through API and bi-directional dynamic data sharing.
  • Automatic configured transition and program driven transition of Voiceflow processing among Voiceflow Modules.
  • Comprehensive real-time Voiceflow processing event notifications to programs.
  • Automatic and configured handling of interruptions from device or other programs.
  • Devide and offline audio recording, speech recognition, speech synthesis and voice activity detection tasks.
  • Audio session and device speaker control ... and much more.

STVoiceFlow framework interprets and processes configured Voiceflow Modules and Audio Prompt Modules. Voiceflow and Audio Prompt modules configurations are expressed using the JSON data structure format and must comply with the Voiceflow JSON schema and the Audio Prompt Module JSON schema associated with these configurations.

STVoiceFlowRunner module processes the following Voiceflow Module types:

  • Play Audio
  • Record Audio
  • Recognize Audio
  • Audio Dialog
  • Audio Listener
  • Pause Resume
  • Process

STVoiceFlow framework provided STVoiceFlowRunner module allows developers to quickly and effectively integrate complex conversational speech interactions into their programs. That adds the additional speech input modality to existing user input modalities such as touch, gesture, mouse, keyboard, etc. Real-time media event notification callbacks from STVoiceFlowRunner module to programs via the STVoiceFlowCallback protocol provides developers with maximum control over the management of STVoiceFlowRunner's asynchronous Voiceflow processing and over the real-time behavior of their programs.

STMediaRunner module provided by STVoiceFlow framework provides developers with a low level and a very powerful media SDK that allows the execution of media tasks such as audio playback, audio recording, speech recognition, voice activity detection, etc. STVoiceFlowRunner module is dependent on STMediaRunner module to execute low level media tasks to support the execution of complex conversational speech interactions. STVoiceFlow framework exposes STMediaRunner module to developers for any necessary low level development and testing. STMediaRunner module posts real-time media event notifications to programs via STEventsObserverDelegate delegate protocol.

Get Started

  • Download STVoiceFlow framework and integrate into your program its STVoiceFlowRunner module to enable conversational speech user interfaces with your program users. You may also integrate STMediaRunner module into your program to perform low level media tasks easily and effectively. (Sample applications which integrate with both modules are coming soon.)
  • Access the reference guide and API documentation for STVoiceFlow Framework.
  • Questions? Our Support page is the best place to reach us.

Requirements

For Apple based applications, STVoiceFlow 2.0.0 requires Xcode 16 and higher, iOS 18.0 and above, visionsOS 2.6 and above, and macOS 15.0 sequoia and above.

For android applications, STVoiceFlow framework is under development.

Licensing

STVoiceFlow framework is available for development at no cost.