# Concepts We Work on

Dhee.AI is a vertically integrated stack of natural language processing and speech technologies which come together to work seamlessly as voice assisted chat-bots and voice bots - capable to concurrently engage with thousands of users.

To help you in training the bots better on our platform, we shall be giving you key insights on the components and concepts which are the building blocks of Dhee.AI.

Equipped with these insights, you shall be able to debug your bots faster and take quick and effective remedial actions to fix the bugs or gaps identified by your end users.

### Natural Language Processing (NLP)

At the heart of what we do is Natural language processing. You can get a short introduction to it by visiting the below link. Also note that the components of NLP used extensively in Dhee.AI are introduced as sub-topics which you can access and read using the corresponding links.

{% content-ref url="/pages/1xKyzAh6zcB4EfyHzn28" %}
[Natural Language Processing (NLP)](/concepts-we-work-on/natural-language-processing-nlp.md)
{% endcontent-ref %}

![](/files/O985XPx7ehtUKH8KNfmp)

#### Natural Language Parsers

The first action done by Dhee.AI when a human speaks to it is to parse it using a natural language parser to get the parts of the speech, entities, and the general semantic relationship between entities and verbs in the spoken utterance.

Read more about this elaborate parsing pipeline using the below link.

{% content-ref url="/pages/Q5L52d37v88ubazHZqWS" %}
[Natural Language Parser Pipeline](/concepts-we-work-on/natural-language-parser-pipeline.md)
{% endcontent-ref %}

#### Context Vectors (Word Embeddings)

Parsing and finding structure of a spoken utterance is only the beginning of the journey. The next step involves semantic analysis. Here we try to enrich the parse output with more semantic information. This is done by injecting the contextual vector of each word into the parse output.

![](/files/avD2xyLmBZAk977yrgwp)

Read more about word embedding by visiting the below link.

{% content-ref url="/pages/PdlHxLGBqOBGdyAMhzvP" %}
[Word Embeddings](/concepts-we-work-on/word-embeddings.md)
{% endcontent-ref %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://readme.dhee.ai/concepts-we-work-on.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.