skills.rest
calesthio
calesthioCOMMUNITY
Explore all skills by this author →
𝕏@calesthioailabsGitHub

azure-speech-to-text

Transcribe audio files to text with word-level timestamps using Azure Fast Transcription API.

55.1k|6.9k|Updated Mar 29, 2026
Search Tags:#subtitles#transcription#Azure#speech-to-text#diarization#REST api#Audio
One-click install
npx skills add https://github.com/calesthio/OpenMontage --skill azure-speech-to-text
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: azure-speech-to-text
Source: https://github.com/calesthio/OpenMontage/tree/main/.agents/skills/azure-speech-to-text
Command: npx skills add https://github.com/calesthio/OpenMontage --skill azure-speech-to-text

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests.

What problem does it solve?

Converting spoken audio or video content into accurate, timestamped text is essential for subtitle generation, content indexing, and video editing pipelines, but setting up cloud transcription often involves complex batch jobs, blob storage, and SDK dependencies. This Skill provides a straightforward path to Azure AI Speech Fast Transcription using only an API key, region, and simple HTTP requests.

Core Features & Use Cases

  • Fast Transcription via REST API: Transcribe local audio files synchronously with sub-real-time latency, no Blob storage or SAS URLs required.
  • Word-Level Timestamps & Diarization: Get per-word timing for subtitle cue generation and optional speaker labels for multi-speaker content like interviews and podcasts.
  • Drop-in Compatibility: Output matches the OpenMontage transcriber schema exactly, making it interchangeable with the local faster-whisper tool for subtitle_gen and downstream stages.
  • Use Case: You have a recorded podcast episode as an MP3 and need subtitles. Configure AZURE_SPEECH_KEY and AZURE_SPEECH_REGION, run azure_stt with diarization enabled, and feed the resulting segments directly into subtitle generation.

Quick Start

Set the AZURE_SPEECH_KEY and AZURE_SPEECH_REGION environment variables, then ask the agent to transcribe your audio file with azure_stt and generate timestamped segments for subtitles.

Frequently Asked Questions about azure-speech-to-text

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I transcribe audio to text with Azure Speech?▼

Set AZURE_SPEECH_KEY and AZURE_SPEECH_REGION environment variables, then call the azure_stt tool with your audio file path. It uses Azure Fast Transcription, a synchronous REST API that returns segments with word-level timestamps without needing Blob storage or SDKs.

Azure Fast Transcription vs Batch Transcription for audio files?▼

Fast Transcription accepts local files via multipart POST and returns results synchronously in sub-real-time, needing only a key and region. Batch Transcription requires audio hosted at a URL with Blob storage and SAS tokens, plus async job polling, suiting bulk or very long jobs.

Does Azure speech-to-text support speaker diarization?▼

Yes, enable the diarize parameter and optionally set max_speakers to get speaker labels per segment. Unlike some local solutions, it requires no HuggingFace token, making it suitable for multi-speaker content like interviews and podcasts.

Can I use Azure transcription offline or without an API key?▼

No, Azure Fast Transcription requires internet access and a configured AZURE_SPEECH_KEY. For offline transcription, fall back to the local faster-whisper transcriber tool, which has an identical execute signature and output schema.

What are the file size limits for Azure Fast Transcription?▼

A single request handles files up to roughly two hours or a few hundred megabytes. For longer recordings or bulk processing, use Azure Batch Transcription instead, and consider transcoding video to audio first to reduce upload size.

Related Skills

Explore More Software Engineering Skills→
cncorpcncorp

register-twilio-test-audio

Upload test audio files to S3 and register URLs in twilio_place_call.py.

Official
Basic
gooseworks-aigooseworks-ai
6

sync-captions-to-music

Generate beat-synced captions for music-video ads with burned-in subtitles.

Official
Advanced
MOODMNKY-LLCMOODMNKY-LLC

openai-whisper

Transcribe local audio files to text using the Whisper CLI.

Community
Basic
Lingtai-AILingtai-AI
621

listen

Transcribes speech and analyzes audio locally using whisper and librosa without network calls.

Official
Intermediate