This Azure AI Services article will show how to create an Azure AI Language Account using REST API.
Table of contents
Azure AI Language allows us to build applications based on language models that can be trained with data. By training a language model, we enable it to understand a user’s meaning when processing input.
With Azure AI Language, we can create applications that are Natural Language Processing (NLP) enabled, allowing them to understand human language in written or spoken form.
Azure AI Language offers two types of service features: pre-configured models, which can be used immediately, and learned (trained) models, which require building and training before use.
Regardless of the feature used, the usage concept is the same. Once a model is provisioned or trained, we access it using a service endpoint and send queries that are processed by the model.
Out-of-the-Box Features
Azure AI Language provides the following pre-configured capabilities that do not require training:
- Summarizing text
- Extracting entities
- Detecting personal information (PII)
- Key phrase extraction
- Sentiment analysis
- Language detection
When not using the pre-configured capabilities and opting to use the learned features, we need to label and train the model on our data before making it available (deploying).
The underlying goal of using Azure AI Language is to develop a Conversational Language Understanding (CLU) model that can predict user intentions.
The most common use case for Azure AI Language is developing and building a Q&A model that is trained on custom data.
How to Create an Azure AI Language Account Using REST API
To create an AI language account, use the following POST request. Make sure you have an App Registration with the right permissions.
Create POST Request
https://management.azure.com/subscriptions/{{subscriptionId}}/resourceGroups/{{RGNAME}}/providers/Microsoft.CognitiveServices/accounts/{{ACCOUNTNAME}}?api-version=2023-05-01Request Body
Add the JSON file below to the request body.
{
"location": "East US",
"kind": "TextAnalytics",
"sku": {
"name": "F0"
}
}
Once created, you will see the account on the Azure portal.
Trackbacks/Pingbacks