Types of language API services
Azure AI Language is a cloud-based service that provides Natural Language Processing (NLP) features for understanding and analyzing text. This service can help you build intelligent applications. It provides tools like the web-based Language Studio, REST APIs, and client libraries.
Services
Here are some details about language API services:
- Azure Cognitive Service for Language provides several NLP features for understanding and analyzing text. This service brings together Text Analytics, QnA Maker, and LUIS. These features can be:
- Preconfigured, which means that the AI models the feature uses aren't customizable. You just send your data and use the feature's output in your applications.
- Customizable, which means that you use Azure AI services tools to train an AI model to fit your data.
- Azure OpenAI Service provides REST API access to powerful OpenAI language models, including GPT-3, Codex, and embeddings. You can easily adopt these models to your specific task. Tasks include content generation, summarization, semantic search, and natural-language-to-code translation. You can access the service via REST APIs, a Python SDK, or the web-based interface in Azure OpenAI Studio.
- Cognitive Services Translator is a translation service that provides text-to-text APIs.
- QnA Maker is a cloud-based NLP service that you can use to create a natural conversational layer over your data.
Use cases
The following table provides recommended services for specific use cases.
Use case | Service to use | Service category |
---|---|---|
Translation | ||
Translate industry-specific text | Cognitive Services Custom Translator | Translator |
Translate generic text that isn't specific to an industry | Cognitive Services Translator | Translator |
Translate natural language into SQL queries | Azure OpenAI natural language to SQL | Azure OpenAI |
Enable apps to process and analyze natural language | Cognitive Services conversational language understanding | Language |
Identification | Language | |
Identify sensitive information and PII | Cognitive Services Personally Identifiable Information (PII) detection | Language |
Identify sensitive information, PII, and PHI | Cognitive Services Text Analytics for health | Language |
Identify entities in text and categorize them into predefined types | Cognitive Services named entity recognition | Language |
Extract domain-specific entities or information | Cognitive Services custom named entity recognition | Language |
Extract the main key phrases from text | Cognitive Services key phrase extraction | Language |
Summarize a document | Azure OpenAI GPT-3 text summarization | Azure OpenAI |
Classification | ||
Classify text by using sentiment analysis | Cognitive Services sentiment analysis | Language |
Classify text by using custom classes | Cognitive Services custom text classification | Language |
Classify items into categories provided at inference time | Azure OpenAI text classification | Azure OpenAI |
Classify text by detecting the language | Cognitive Services language detection | Language |
Understanding context | ||
Link an entity with knowledge base articles | Cognitive Services Entity Linking | Language |
Understand questions and answers (generic) | Cognitive Services question answering | QnA Maker |
Understand questions and answers (custom) | Cognitive Services custom question answering | QnA Maker |
Combine multiple cognitive services | Cognitive Services orchestration workflow | Language |
Contributors
This article is maintained by Microsoft. It was originally written by the following contributors.
Principal authors:
- Ashish Chahuan | Senior Cloud Solution Architect
- Kruti Mehta | Azure Senior Fast-Track Engineer
Other contributors:
- Mick Alberts | Technical Writer
- Brandon Cowen | Senior Cloud Solution Architect
- Oscar Shimabukuro | Senior Cloud Solution Architect
- Manjit Singh | Software Engineer
- Christina Skarpathiotaki | Senior Cloud Solution Architect
- Nathan Widdup | Azure Senior Fast-Track Engineer
To see nonpublic LinkedIn profiles, sign in to LinkedIn.
Next steps
- What is Azure Cognitive Service for Language?
- Language APIs blog post
- Learning path: Create a language understanding solution with Azure Cognitive Services
- Learning path: Provision and manage Azure Cognitive Services
- Learning path: Identify principles and practices for responsible AI
- Learning path: Introduction to responsible bots