Fun with DSL
Download as PPTX, PDF1 like354 views
The document outlines the creation of a Cortana/Siri-like assistant using domain-specific languages (DSL). It emphasizes the components needed, such as a user interface, grammar, parser, and translator, and discusses tools like Irony.NET for building DSLs. Furthermore, it distinguishes between speech recognition and voice recognition while detailing implementation methods for web and mobile applications.
Fun with DSL
- 1. with Domain Specific Languages How to create a Cortana / Siri like assistant for our applications Pedro Ramírez Scio Consulting
- 3. The pieces • A UI that recognice what the user said or free textbox entry • A language that the user uses and the assistant understand • A grammar • A parser • A translator
- 4. What is a DSL? How to implement a DSL? There are many tools to build DLSs, for .Net some of them are: • Visual Studio • Irony.Net • Boo language • ANTLR For this demo I use Irony.Net, Irony.net allows us to define the grammar and provides a parser that we use to then translate the expression entered by the user in to a SQL query. A domain-specific language (DSL) is a computer language specialized to a particular application domain. This is in contrast to a general-purpose language (GPL), which is broadly applicable across domains, and lacks specialized features for a particular domain.
- 5. Voice or Speech recognition? How to implement what we need? For our purposes, we need speech recognition rather than voice recognition, we need to translate the user words into text that can be used by the parser so we can “understand” what the user said. We can implement speech recognition in a few ways, for web apps the two most common are: • Using native windows libraries (System.Speech), this is can be used by desktop and web apps. • For web applications(our case), we can use webkitSpeechRecognition in javascript Mobile applications can use native libraries, like the ones for Windows Phone 8 Speechrecognition is the translation of spoken words into text. Voice recognition refers to identify the speaker, not what they are saying.
- 7. Questions? Useful links: Source code https://github.com/pedro-ramirez-suarez/ScioAssistant Irony .Net https://irony.codeplex.com/