Nuance Speech Suite
Speech Suite is a package of components known as Nuance speech products. Each component has one or more services and requires a license: the license determines which components you can use, which features of those components, and the amount of usage allowed. Components:
-
Nuance Speech Server: Central control and communication hub for speech-processing resources. Speech Server provides an open, protocol-based mechanism (MRCP) for voice platforms to issue recognition and audio output (prerecorded or text-to-speech) requests. See Speech Server features.
Note: Nuance strongly recommends using MRCPv2. Although Speech Server also supports both MRCPv1, other components do not. (This documentation generally describes MRCPv2 usage, and does not always include details about MRCPv1.)
-
Nuance Recognizer: Recognition engine for grammar-based, constrained vocabularies (“Please say yes or no.” or "What is the destination city?"), and for statistical models that allow natural speech in response to open-ended questions (“Hello. How can I help you today?”). See Recognizer features.
-
Dragon Voice: Recognition engines for transcription and open-dialog (virtual assistant) applications. See Dragon Voice features.
-
Nuance Vocalizer for Enterprise: Text-to-speech (TTS) engine for speaking to customers. See Vocalizer features.
-
Nuance License Manager: Required server accessed by all Speech Suite components.
-
Nuance Resource Manager: Required server for on-premise Dragon Voice engines. It reserves resources and balances loads.
-
Nuance Management Station: Optional tool for configuring, deploying, administering, and managing Speech Suite services.
Details about Dragon Voice engines
Dragon Voice is a component in every Speech Products installation, and you can use it if you acquire the needed license. It includes these sub-components:
-
Natural Language Processing service: Middleware component that manages the connections (WebSocket) between Speech Server and Dragon Voice engines.
-
Krypton recognition engine: Engine for real-time recognition. Supports raw recognition and open-dialog (large vocabulary, continuous) recognition. The Krypton engine uses a WebSocket-based protocol to accept audio streams and asynchronously return transcription results as the recognition progresses. Requires data packs to remain current with popular vocabulary. (Download the packs from Nuance Network.)
-
Natural Language Engine: Engine for semantic processing and meaning extraction. Requires domain language models (DLMs) to recognize and understand the specific terminology of your business or environment. To create DLMs, use Nuance Command Line Interface or Nuance Experience Studio.
-
Nuance Text Processing Engine: Engine for normalization and tokenization (lexical analysis). Requires data packs to remain current with popular vocabulary. (Download the packs from Nuance Network.) Also requires domain language models (DLMs) to recognize and understand the specific terminology of your business or environment. To create DLMs, use Nuance Command Line Interface or Nuance Experience Studio.
Related products
These Nuance products are not part of the Speech Suite package, but you can use them with Speech Suite solutions:
- Nuance Command Line Interface: Tool for creating Dynamic Language Models (DLMs) and models for Natural Language Understanding (NLU).
- Nuance Experience Studio: Web-based tool for creating NLU (natural language understanding) models for virtual assistant and call steering applications. These models enable the application to understand what users mean when they contact the application. Dragon Voice applications rely on NLU models (domain language models for recognition, and semantic models for interpretation) that are generated via Nuance Experience Studio.
- Nuance Mix Tools: Web-based tool for creating NLU (natural language understanding) models and dialogs.
- Nuance Application Studio: Web-based tool for designing and developing speech and touchtone applications. Nuance Application Studio streamlines the design process, facilitates communication with stakeholders, and generates code and other collateral.
- Nuance Insights for IVR: Analysis and reporting tool that provides usage information about speech, mobile, and touchtone applications based on call logs and audio files collected from applications.