Azure’s Natural Language Processing Services (NLP)
→ In this article, we will understand high-level multiple NLP services overview.
→ With NLP services, we get intelligence from a conversation, speech, or written text in readable language.
→ This is part of Azure Cognitive Service.
A. Language:
Language: Extracting meaning from unstructured text (in NLP)
1. Text Analytics API:
→ This service detects — sentiments, key phrases, and named entities.
STEP 1: We write some text. Click the Next button.
STEP 2: Text Analytics will give us below details:
a. Sentiment — Neutral/ Negative /Positive
b. Highlights the keys (as shown in blue color text above)
→ In the next step, it also gives — entity linking. Let us see below:
Entity Linking: Opening adding wikipedia link in highlighted text
→ If we click the highlighted link
STEP 4: In the next step we have to search entity from Bing search engine.
→ Basically fetched Bing search results in the web app
Eg 2: Let us try with another text, Text: I hate it
→ We observe detected sentiment is — negative.
2. Translator API:
→ This service has the ability to translate to or from 90 languages. Check more here.
Speech:
→ We use the below service to integrate Speech into the apps and services:
3. Speech Service API:
→ This service handles — Speech to Text, Text to Speech, Translation and Speaker Recognition. Check out more about this service here.
4. QnA Maker API (under Build Conversation):
→ If we want to make a chatbot on our website. Ask some set of questions and get answers from the bot itself we will make use of this service.
→ It helps to make a conversational question and answer layer. Check more here.
5. LUIS API (Language Understanding Intelligent Service):
→ LUIS understands spoken or text commands.
→ It also understands the user’s natural language utterance (i.e. human voice) somewhat Alexa device uses.
Eg.
Turn lights on,
Turn lights off,
Book me a flight to Japan,
Order two KFC Chicken Buckets 😅
→ So basically it understands our speech/actions and performs some actions.
→ Let us check with an example (shared on the Microsoft website)
On Left some command buttons, text input and speech microphone, and
On the right four rooms with lights
Eg 1: Click Go to energy saver mode. We observe the light is switched off.
Eg 2: Click on Concentrating on my work button. Specific light turns on.
Eg 3: We can also write text for command — Turn lights off and click Apply button. We observe all lights turned off.
Eg 4: We can also use the Speech icon and LUIS will execute for us.
→ Know more about LUIS here.
Closing Thoughts:
In this article, we have learned a high-level overview of Azure’s Natural Language Processing (NLP) services.
There are basically five AI services — Text Analytics, Translator, Speech Service, QnA Maker, and Language Understanding Intelligent Services (LUIS) APIs with specific AI work in applications.
Thank you for reading till the end 🙌 . If you enjoyed this article or learned something new, support me by clicking the share button below to reach more people and/or give me a follow on Twitter and subscribe Happy Learnings !! to see some other tips, articles, and things I learn about and share there.
