Azure Cognitive Services: AI APIs for Every Developer

Azure Cognitive Services bring AI capabilities to developers without ML expertise. Pre-built APIs for vision, speech, language, and decision.

Service Categories

Vision

• Computer Vision: Image analysis, OCR
• Face: Detection, recognition, emotions
• Custom Vision: Train your own image classifier

Speech

• Speech to Text: Transcription
• Text to Speech: Voice synthesis
• Speech Translation: Real-time translation

Language

• Text Analytics: Sentiment, key phrases, entities
• Translator: 70+ languages
• LUIS: Language understanding

Decision

• Anomaly Detector: Time series anomalies
• Content Moderator: Image/text moderation
• Personalizer: Personalized recommendations

Quick Start: Text Analytics

from azure.ai.textanalytics import TextAnalyticsClient
from azure.core.credentials import AzureKeyCredential

client = TextAnalyticsClient(
    endpoint="https://my-cog.cognitiveservices.azure.com/",
    credential=AzureKeyCredential("your-key")
)

documents = ["I love this product! It's amazing.", "Terrible service, very disappointed."]

# Sentiment analysis
results = client.analyze_sentiment(documents)
for doc in results:
    print(f"Sentiment: {doc.sentiment}, Scores: {doc.confidence_scores}")

# Key phrases
key_phrases = client.extract_key_phrases(documents)
for doc in key_phrases:
    print(f"Key phrases: {doc.key_phrases}")

Computer Vision

from azure.cognitiveservices.vision.computervision import ComputerVisionClient

client = ComputerVisionClient(endpoint, CognitiveServicesCredentials(key))

# Analyze image
analysis = client.analyze_image(image_url, visual_features=['Categories', 'Description', 'Tags'])

print(f"Description: {analysis.description.captions[0].text}")
print(f"Tags: {[tag.name for tag in analysis.tags]}")

Pricing Model

Most services offer:

• Free tier: Limited calls/month
• Standard tier: Pay per transaction

Example: Text Analytics sentiment = $0.50 per 1,000 text records

Cognitive Services democratize AI for application developers.