Azure AI Transcription Py

1---
2name: azure-ai-transcription-py
3description: |
4  Azure AI Transcription SDK for Python. Use for real-time and batch speech-to-text transcription with timestamps and diarization.
5  Triggers: "transcription", "speech to text", "Azure AI Transcription", "TranscriptionClient".
6package: azure-ai-transcription
7---
8 
9# Azure AI Transcription SDK for Python
10 
11Client library for Azure AI Transcription (speech-to-text) with real-time and batch transcription.
12 
13## Installation
14 
15```bash
16pip install azure-ai-transcription
17```
18 
19## Environment Variables
20 
21```bash
22TRANSCRIPTION_ENDPOINT=https://<resource>.cognitiveservices.azure.com
23TRANSCRIPTION_KEY=<your-key>
24```
25 
26## Authentication
27 
28Use subscription key authentication (DefaultAzureCredential is not supported for this client):
29 
30```python
31import os
32from azure.ai.transcription import TranscriptionClient
33 
34client = TranscriptionClient(
35    endpoint=os.environ["TRANSCRIPTION_ENDPOINT"],
36    credential=os.environ["TRANSCRIPTION_KEY"]
37)
38```
39 
40## Transcription (Batch)
41 
42```python
43job = client.begin_transcription(
44    name="meeting-transcription",
45    locale="en-US",
46    content_urls=["https://<storage>/audio.wav"],
47    diarization_enabled=True
48)
49result = job.result()
50print(result.status)
51```
52 
53## Transcription (Real-time)
54 
55```python
56stream = client.begin_stream_transcription(locale="en-US")
57stream.send_audio_file("audio.wav")
58for event in stream:
59    print(event.text)
60```
61 
62## Best Practices
63 
641. **Enable diarization** when multiple speakers are present
652. **Use batch transcription** for long files stored in blob storage
663. **Capture timestamps** for subtitle generation
674. **Specify language** to improve recognition accuracy
685. **Handle streaming backpressure** for real-time transcription
696. **Close transcription sessions** when complete
70