azure-ai-transcription-py – skill DevOps & Hạ tầng cho agent AI

Azure AI Speech Transcription for Python

When to Use

• Transcribe an authorized audio file with the azure-ai-transcription client.
• Review timestamped output and options supported by the installed SDK version.
• For streaming or asynchronous batch jobs, select the appropriate Speech API instead.

Prerequisites

Use a project-pinned SDK in an isolated Python environment. Confirm the Speech

resource, endpoint, permissions, audio consent, cost and retention before upload.

Do not log keys, audio contents, or full transcripts by default.

File Example

import os
from azure.core.credentials import AzureKeyCredential
from azure.ai.transcription import TranscriptionClient
from azure.ai.transcription.models import TranscriptionContent, TranscriptionOptions

client = TranscriptionClient(
    endpoint=os.environ["TRANSCRIPTION_ENDPOINT"],
    credential=AzureKeyCredential(os.environ["TRANSCRIPTION_KEY"]),
)
with client:
    with open("audio.wav", "rb") as audio:
        result = client.transcribe(TranscriptionContent(
            definition=TranscriptionOptions(locales=["en-US"]), audio=audio,
        ))
# Inspect result locally; handle empty output and service errors explicitly.

Authentication and API Boundary

DefaultAzureCredential is also supported when the identity has the required

resource role. The previous begin_transcription and begin_stream_transcription

examples did not match this client. Do not invent a streaming adapter.

Example Request

“Transcribe this consented local recording; keep the transcript private.”

Limitations

Verify installed signatures and service limits against [Microsoft's SDK guide](https://learn.microsoft.com/en-us/python/api/overview/azure/ai-transcription-readme?view=azure-python).

This illustration has not contacted Azure; availability, accuracy and costs require

verification in the authorized environment.