Natural language text analysis in android application, are there any ready-made solutions?

Question

Java

Hello!)
Now I am studying the issue of natural language processing and knowledge extraction from unstructured data. I want to make some kind of voice assistant for one android application. The user speaks a phrase in natural language, the assistant understands it and executes the commands. As for natural language processing on computers, everything is clear here, a lot of research has already been done and AOT systems have been created, combining which you can achieve good results.
But is there something for android? After all, we have big limitations on smartphones: memory, computing capabilities, mobile Internet, etc. Of all the possible options, I have so far found only one - api.ai. It seems to provide natural language processing capabilities, you can describe your subject area, but everything rests on the need for a constant connection to the Internet. Now this is certainly not an acute problem, but still a problem.
If there are no ready-made libraries, then maybe there are some studies on this subject, algorithms.
What are you interested in:

1. How to perform parsing?
   
2. How to perform semantic analysis?
   
3. How to describe and where to store the knowledge base?
   

Answer 1

Yandex has a speech processing library - Yandex Speech Kit .
There is also PortableSphinx - there is a voice recognizer, a library, but you have to do the semantic (semantic) analysis of speech yourself.
So there is, if you are satisfied with the second option - look for PortableSphinx competitors, there are about 2-3 opensource libraries for voice recognition.
Now closer to the answer to your actual question - that's what I would do through machine learning. That is, training the network in a specific language based on incoming data - for example, to recognize a single voice. I am not strong in this topic, but I think this is just what you need.
Afterlinks: 1 , 2 , 3