Use Azure AI Language text analytics in Fabric with REST API and SynapseML (preview)

2025-07-31

Important

This feature is in preview.

Azure AI Language is an Azure AI service that enables you to perform text mining and text analysis with Natural Language Processing (NLP) features.

In this article, you learn how to use Azure AI Language services directly in Microsoft Fabric to analyze text. By the end of this article, you are able to:

Detect sentiment labels at the sentence or document level
Identify the language for a given text input
Extract key phrases from a text
Identify different entities in text and categorize them into predefined classes or types

Prerequisites

Get a Microsoft Fabric subscription. Or, sign up for a free Microsoft Fabric trial.
Sign in to Microsoft Fabric.
Use the experience switcher on the bottom left side of your home page to switch to Fabric.

Create a new notebook.
Attach your notebook to a lakehouse. On the left side of your notebook, select Add to add an existing lakehouse or create a new one.

Note

This article uses Microsoft Fabric's built-in prebuilt AI services, which handle authentication automatically. You don't need to obtain a separate Azure AI services key - the authentication is managed through your Fabric workspace. For more information, see Prebuilt AI models in Fabric (preview).

The code samples in this article use libraries that are preinstalled in Microsoft Fabric notebooks:

SynapseML: Preinstalled in Fabric notebooks for machine learning capabilities
PySpark: Available by default in Fabric Spark compute
Standard Python libraries: json, uuid are part of Python standard library

Note

Microsoft Fabric notebooks come with many common libraries preinstalled. The SynapseML library, which provides the MLflow integration and text analytics capabilities, is automatically available in the Spark environment.

Choose your approach

This article provides two ways to use Azure AI Language services in Fabric:

REST API approach: Direct HTTP calls to the service (recommended for beginners)
SynapseML approach: Using Spark DataFrames for larger-scale processing

Tip

New users should start with the REST API approach as it's easier to understand and debug. The SynapseML approach is better for processing large datasets with Spark.

Rest API
SynapseML

Set up authentication and endpoints

Copy and paste this code into the first cell of your Fabric notebook to set up the connection to Azure AI Language services:

Note

This code uses Fabric's built-in authentication. The get_fabric_env_config function automatically retrieves your workspace credentials and connects to the prebuilt AI services. No API key is required.

# Get workload endpoints and access token
from synapse.ml.fabric.service_discovery import get_fabric_env_config
from synapse.ml.fabric.token_utils import TokenUtils
import json
import requests

fabric_env_config = get_fabric_env_config().fabric_env_config
auth_header = TokenUtils().get_openai_auth_header()

# Make a RESful request to AI service
prebuilt_AI_base_host = fabric_env_config.ml_workload_endpoint + "cognitive/textanalytics/"
print("Workload endpoint for AI service: \n" + prebuilt_AI_base_host)

service_url = prebuilt_AI_base_host + "language/:analyze-text?api-version=2022-05-01"
print("Service URL: \n" + service_url)

auth_headers = {
    "Authorization" : auth_header
}

def print_response(response):
    if response.status_code == 200:
        print(json.dumps(response.json(), indent=2))
    else:
        print(f"Error: {response.status_code}, {response.content}")

Import required libraries

Copy and paste this code into the first cell of your Fabric notebook:

import synapse.ml.core
from synapse.ml.cognitive.language import AnalyzeText
from pyspark.sql.functions import col

# Note: 'spark' and 'display()' are automatically available in Fabric notebooks

The Sentiment Analysis feature provides a way for detecting the sentiment labels (such as "negative", "neutral" and "positive") and confidence scores at the sentence and document-level. This feature also returns confidence scores between 0 and 1 for each document and sentences within it for positive, neutral, and negative sentiment. See the Sentiment Analysis and Opinion Mining language support for the list of enabled languages.

Analyze sentiment of text

Copy this code into a new cell in your notebook to analyze the sentiment of a sample text:

payload = {
    "kind": "SentimentAnalysis",
    "parameters": {
        "modelVersion": "latest",
        "opinionMining": "True"
    },
    "analysisInput":{
        "documents":[
            {
                "id":"1",
                "language":"en",
                "text": "The food and service were unacceptable. The concierge was nice, however."
            }
        ]
    }
} 

response = requests.post(service_url, json=payload, headers=auth_headers)


# Output all information of the request process
print_response(response)

Tip

You can replace the text in the "text" field with your own content to analyze. The service returns sentiment scores and identify which parts of the text are positive, negative, or neutral.

Expected output

When you run the following code successfully, you should see output similar to this:

{
  "kind": "SentimentAnalysisResults",
  "results": {
    "documents": [
      {
        "id": "1",
        "sentiment": "negative",
        "confidenceScores": {
          "positive": 0.0,
          "neutral": 0.0,
          "negative": 1.0
        },
        "sentences": [
          {
            "sentiment": "negative",
            "confidenceScores": {
              "positive": 0.0,
              "neutral": 0.0,
              "negative": 1.0
            },
            "offset": 0,
            "length": 40,
            "text": "The food and service were unacceptable. ",
            "targets": [
              {
                "sentiment": "negative",
                "confidenceScores": {
                  "positive": 0.01,
                  "negative": 0.99
                },
                "offset": 4,
                "length": 4,
                "text": "food",
                "relations": [
                  {
                    "relationType": "assessment",
                    "ref": "#/documents/0/sentences/0/assessments/0"
                  }
                ]
              },
              {
                "sentiment": "negative",
                "confidenceScores": {
                  "positive": 0.01,
                  "negative": 0.99
                },
                "offset": 13,
                "length": 7,
                "text": "service",
                "relations": [
                  {
                    "relationType": "assessment",
                    "ref": "#/documents/0/sentences/0/assessments/0"
                  }
                ]
              }
            ],
            "assessments": [
              {
                "sentiment": "negative",
                "confidenceScores": {
                  "positive": 0.01,
                  "negative": 0.99
                },
                "offset": 26,
                "length": 12,
                "text": "unacceptable",
                "isNegated": false
              }
            ]
          },
          {
            "sentiment": "neutral",
            "confidenceScores": {
              "positive": 0.22,
              "neutral": 0.75,
              "negative": 0.04
            },
            "offset": 40,
            "length": 32,
            "text": "The concierge was nice, however.",
            "targets": [
              {
                "sentiment": "positive",
                "confidenceScores": {
                  "positive": 1.0,
                  "negative": 0.0
                },
                "offset": 44,
                "length": 9,
                "text": "concierge",
                "relations": [
                  {
                    "relationType": "assessment",
                    "ref": "#/documents/0/sentences/1/assessments/0"
                  }
                ]
              }
            ],
            "assessments": [
              {
                "sentiment": "positive",
                "confidenceScores": {
                  "positive": 1.0,
                  "negative": 0.0
                },
                "offset": 58,
                "length": 4,
                "text": "nice",
                "isNegated": false
              }
            ]
          }
        ],
        "warnings": []
      }
    ],
    "errors": [],
    "modelVersion": "2025-01-01"
  }
}

The Sentiment Analysis feature provides a way for detecting the sentiment labels (such as "negative", "neutral" and "positive") and confidence scores at the sentence and document-level. This feature also returns confidence scores between 0 and 1 for each document & sentences within it for positive, neutral, and negative sentiment. See the Sentiment Analysis and Opinion Mining language support for the list of enabled languages.

df = spark.createDataFrame([
    ("Great atmosphere. Close to plenty of restaurants, hotels, and transit! Staff are friendly and helpful.",),
    ("What a sad story!",)
], ["text"])

model = (AnalyzeText()
        .setTextCol("text")
        .setKind("SentimentAnalysis")
        .setOutputCol("response"))

result = model.transform(df)\
        .withColumn("documents", col("response.documents"))\
        .withColumn("sentiment", col("documents.sentiment"))

display(result.select("text", "sentiment"))

Language detector

Rest API
SynapseML

The Language Detector evaluates text input for each document and returns language identifiers with a score that indicates the strength of the analysis. This capability is useful for content stores that collect arbitrary text, where language is unknown. See the Supported languages for language detection for the list of enabled languages.

payload = {
    "kind": "LanguageDetection",
    "parameters": {
        "modelVersion": "latest"
    },
    "analysisInput":{
        "documents":[
            {
                "id":"1",
                "text": "This is a document written in English."
            }
        ]
    }
}

response = requests.post(service_url, json=payload, headers=auth_headers)

# Output all information of the request process
print_response(response)

Output

{
  "kind": "LanguageDetectionResults",
  "results": {
    "documents": [
      {
        "id": "1",
        "warnings": [],
        "detectedLanguage": {
          "name": "English",
          "iso6391Name": "en",
          "confidenceScore": 0.95
        }
      }
    ],
    "errors": [],
    "modelVersion": "2024-11-01"
  }
}

df = spark.createDataFrame([
    (["Hello world"],),
    (["Bonjour tout le monde", "Hola mundo", "Tumhara naam kya hai?"],),
    (["你好"],),
    (["日本国（にほんこく、にっぽんこく、英"],)
], ["text"])

model = (AnalyzeText()
        .setTextCol("text")
        .setKind("LanguageDetection")
        .setOutputCol("response"))

result = model.transform(df)\
        .withColumn("documents", col("response.documents"))\
        .withColumn("detectedLanguage", col("documents.detectedLanguage.name"))

display(result.select("text", "detectedLanguage"))

Key Phrase Extractor

Rest API
SynapseML

The Key Phrase Extraction evaluates unstructured text and returns a list of key phrases. This capability is useful if you need to quickly identify the main points in a collection of documents. See the Supported languages for key phrase extraction for the list of enabled languages.

payload = {
    "kind": "KeyPhraseExtraction",
    "parameters": {
        "modelVersion": "latest"
    },
    "analysisInput":{
        "documents":[
            {
                "id":"1",
                "language":"en",
                "text": "Dr. Smith has a very modern medical office, and she has great staff."
            }
        ]
    }
}

response = requests.post(service_url, json=payload, headers=auth_headers)

# Output all information of the request process
print_response(response)

Output

{
  "kind": "KeyPhraseExtractionResults",
  "results": {
    "documents": [
      {
        "id": "1",
        "keyPhrases": [
          "modern medical office",
          "Dr. Smith",
          "great staff"
        ],
        "warnings": []
      }
    ],
    "errors": [],
    "modelVersion": "2022-10-01"
  }
}

df = spark.createDataFrame([
    ("en", "Microsoft was founded by Bill Gates and Paul Allen."),
    ("en", "Text Analytics is one of the Azure Cognitive Services."),
    ("en", "My cat might need to see a veterinarian.")
], ["language", "text"])

model = (AnalyzeText()
        .setTextCol("text")
        .setKind("KeyPhraseExtraction")
        .setOutputCol("response"))

result = model.transform(df)\
        .withColumn("documents", col("response.documents"))\
        .withColumn("keyPhrases", col("documents.keyPhrases"))

display(result.select("text", "keyPhrases"))

Named Entity Recognition (NER)

Rest API
SynapseML

Named Entity Recognition (NER) is the ability to identify different entities in text and categorize them into predefined classes or types such as: person, location, event, product, and organization. See the NER language support for the list of enabled languages.

payload = {
    "kind": "EntityRecognition",
    "parameters": {
        "modelVersion": "latest"
    },
    "analysisInput":{
        "documents":[
            {
                "id":"1",
                "language": "en",
                "text": "I had a wonderful trip to Seattle last week."
            }
        ]
    }
}

response = requests.post(service_url, json=payload, headers=auth_headers)

# Output all information of the request process
print_response(response)

Output

{
  "kind": "EntityRecognitionResults",
  "results": {
    "documents": [
      {
        "id": "1",
        "entities": [
          {
            "text": "trip",
            "category": "Event",
            "offset": 18,
            "length": 4,
            "confidenceScore": 0.66
          },
          {
            "text": "Seattle",
            "category": "Location",
            "subcategory": "City",
            "offset": 26,
            "length": 7,
            "confidenceScore": 1.0
          },
          {
            "text": "last week",
            "category": "DateTime",
            "subcategory": "DateRange",
            "offset": 34,
            "length": 9,
            "confidenceScore": 1.0
          }
        ],
        "warnings": []
      }
    ],
    "errors": [],
    "modelVersion": "2025-02-01"
  }
}

df = spark.createDataFrame([
    ("en", "Microsoft was founded by Bill Gates and Paul Allen."),
    ("en", "Pike place market is my favorite Seattle attraction.")
], ["language", "text"])

model = (AnalyzeText()
        .setTextCol("text")
        .setKind("EntityRecognition")
        .setOutputCol("response"))

result = model.transform(df)\
        .withColumn("documents", col("response.documents"))\
        .withColumn("entityNames", col("documents.entities.text"))

display(result.select("text", "entityNames"))

Entity linking

Rest API
SynapseML

No steps for REST API in this section.

Entity linking identifies and disambiguates the identity of entities found in text. For example, in the sentence "We went to Seattle last week.", the word "Seattle" would be identified, with a link to more information on Wikipedia. See Supported languages for entity linking for the list of enabled languages.

df = spark.createDataFrame([
    ("en", "Microsoft was founded by Bill Gates and Paul Allen."),
    ("en", "Pike place market is my favorite Seattle attraction.")
], ["language", "text"])

model = (AnalyzeText()
        .setTextCol("text")
        .setKind("EntityLinking")
        .setOutputCol("response"))

result = model.transform(df)\
        .withColumn("documents", col("response.documents"))\
        .withColumn("entityNames", col("documents.entities.name"))

display(result)

Use prebuilt Text Analytics in Fabric with SynapseML
Use prebuilt Azure AI Translator in Fabric with REST API
Use prebuilt Azure AI Translator in Fabric with SynapseML
Use prebuilt Azure OpenAI in Fabric with REST API
Use prebuilt Azure OpenAI in Fabric with Python SDK
Use prebuilt Azure OpenAI in Fabric with SynapseML
SynapseML GitHub repository - Source code and documentation for SynapseML
Azure AI Language documentation - Complete reference for Azure AI Language service

Share via

Use Azure AI Language text analytics in Fabric with REST API and SynapseML (preview)

Prerequisites

Choose your approach

Set up authentication and endpoints

Sentiment analysis

Analyze sentiment of text

Expected output

Language detector

Output

Key Phrase Extractor

Output

Named Entity Recognition (NER)

Output

Entity linking

Related content

Feedback

Additional resources