1# Azure AI Transcription — Python SDK Quick Reference
2 
3> Condensed from **azure-ai-transcription-py**. Full patterns (real-time streaming, diarization, timestamps)
4> in the **azure-ai-transcription-py** plugin skill if installed.
5 
6## Install
7```bash
8pip install azure-ai-transcription
9```
10 
11## Quick Start
12```python
13import os
14from azure.ai.transcription import TranscriptionClient
15client = TranscriptionClient(endpoint=os.environ["TRANSCRIPTION_ENDPOINT"],
16    credential=os.environ["TRANSCRIPTION_KEY"])
17```
18 
19## Non-Obvious Patterns
20- Auth uses subscription key string directly (not AzureKeyCredential); DefaultAzureCredential not supported
21- Batch: `client.begin_transcription(name=..., locale="en-US", content_urls=[...], diarization_enabled=True)`
22- Real-time: `stream = client.begin_stream_transcription(locale="en-US"); stream.send_audio_file("audio.wav")`
23 
24## Best Practices
251. Enable diarization when multiple speakers are present
262. Use batch transcription for long files stored in blob storage
273. Capture timestamps for subtitle generation
284. Specify language to improve recognition accuracy
295. Handle streaming backpressure for real-time transcription
306. Close transcription sessions when complete
31