Azure AI Speech to Text

Best for: Prototypes and side projects - free to start, no sales call; Regulated or enterprise workloads - compliance attestations and an enterprise plan; AI agents and automation - an agent-ready surface (MCP / llms.txt)

Pricing model

Usage-based [2]blocksentient.com/review/microsoft-azure-speech-service/“Standard Real-time: $1/hour. Fast Transcription: $0.36/hour. Batch Transcription: $0.18/hour. Pay-as-You-Go Rates with no mandatory recurring base fee.”azure.microsoft.com/en-us/pricing/details/speech/

Published pricing

✓ Yes [3]azure.microsoft.com/en-us/pricing/details/speech/

Free tier

✓ Yes [4]blocksentient.com/review/microsoft-azure-speech-service/“The F0 tier provides users with 5 audio hours free per month for both standard and custom speech-to-text (batch excluded), plus one hosted custom model monthly with automatic decommissioning after 7 days if unused.”learn.microsoft.com/en-us/azure/ai-services/speech-service/speech-services-quotas-and-limits“For the Free (F0) pricing tier, see the monthly allowances on the pricing page.”

Free tier details

Free (F0) tier: 5 audio hours per month for Standard and Custom Speech to Text (shared; batch transcription not available on F0). Resets monthly. Concurrent request limit of 1 (not adjustable). No SLA on F0 tier.

Self-serve signup

✓ Yes

Requires sales call

✗ No

Enterprise plan

✓ Yes

• Real-time streaming
• Speaker diarization
• Speech translation

Supported actions

transcribe_realtime, transcribe_batch, transcribe_fast, speaker_diarization, language_detection, word_timestamps, custom_speech_model, pronunciation_assessment, phrase_lists, speech_translation, keyword_recognition, llm_speech_transcription, post_stream_refinement [5]learn.microsoft.com/en-us/azure/ai-services/speech-service/speech-to-text“The speech to text service offers the following core features: Real-time transcription: Instant transcription with intermediate results for live audio inputs. Fast transcription: Fastest synchronous output for situations with predictable latency. Batch transcription: Efficient processing for large volumes of prerecorded audio. Custom speech: Models with enhanced accuracy for specific domains and conditions.”learn.microsoft.com/en-us/azure/ai-services/speech-service/speech-to-text“Diarization is the process of distinguishing and separating different speakers in an audio recording. This feature is particularly useful for transcribing conversations, meetings, or any multi-speaker audio content. The Speech service can identify up to 35 different speakers in an audio recording.”

Regions

South Africa North, East Asia, Southeast Asia, Australia East, Central India, Japan East, Japan West, Korea Central, Canada Central, Canada East, North Europe, West Europe, France Central, Germany West Central, Italy North, Norway East, Sweden Central, Switzerland North, Switzerland West, UK South, UK West, UAE North, Brazil South, Qatar Central, Central US, East US, East US 2, North Central US, South Central US, West Central US, West US, West US 2, West US 3 [6]learn.microsoft.com/en-us/azure/ai-services/speech-service/regions“Azure Speech allows your application to convert audio to text, perform speech translation, and convert text to speech. Azure Speech is available in multiple regions with unique endpoints for the Speech SDK and REST APIs.”

Languages

Afrikaans (South Africa), Amharic (Ethiopia), Arabic (20+ locales), Assamese (India), Azerbaijani, Bulgarian, Bhojpuri (India), Bengali (India), Bosnian, Catalan, Czech, Welsh, Danish, German (3 locales), Greek, English (15+ locales), Spanish (22 locales), Estonian, Basque, Persian, Finnish, Filipino, French (4 locales), Irish, Galician, Gujarati, Hebrew, Hindi, Croatian, Hungarian, Armenian, Indonesian, Icelandic, Italian (2 locales), Japanese, Javanese, Georgian, Kazakh, Khmer, Kannada, Korean, Lao, Lithuanian, Latvian, Macedonian, Malayalam, Mongolian, Marathi, Malay, Maltese, Burmese, Norwegian Bokmål, Nepali, Dutch (2 locales), Odia, Punjabi, Polish, Pashto, Portuguese (2 locales), Romanian, Russian, Sinhala, Slovak, Slovenian, Somali, Albanian, Serbian, Swedish, Kiswahili, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu, Uzbek, Vietnamese, Chinese Wu (Simplified), Chinese Cantonese (Simplified), Chinese Mandarin (Simplified), Chinese Southwestern Mandarin, Chinese Cantonese (Traditional), Chinese Taiwanese Mandarin, isiZulu - 130+ languages/locales total [7]learn.microsoft.com/en-us/azure/ai-services/speech-service/language-support“Speech-to-Text supports 130+ languages and locales.”learn.microsoft.com/en-us/azure/ai-services/speech-service/language-support“Last Updated: June 19, 2026 (per document metadata). Multiple new MAI (Multi-Accent Intelligence) models have been added.”

Input types

audio/wav (PCM, default), audio/mp3, audio/ogg (OPUS), audio/flac, AMR, AMR-WB, A-Law, Mu-Law, streaming via WebSocket (Speech SDK), file via Azure Blob Storage SAS URI, file via public URI, live microphone stream

Output types

JSON (with word-level timestamps, offset, duration, speaker labels), plain text, SRT/VTT captions (via post-processing), word-level timestamps (batch and real-time SDK), diarization speaker labels

Webhooks

✓ Yes [8]learn.microsoft.com/en-us/azure/ai-services/speech-service/batch-transcription-create“Instead of polling for transcription status, you can register a webhook to receive a notification when a transcription job completes (or reaches any other terminal state). The Speech service sends HTTP POST callbacks to your endpoint for transcription.created, transcription.processing, transcription.succeeded, transcription.failed, and transcription.deleted events.”

Sandbox / test mode

✗ No [9]learn.microsoft.com/en-us/azure/ai-services/speech-service/overview“Run Azure Speech anywhere, in the cloud or at the edge in containers. Enable your applications, tools, and devices for speech by using the Speech CLI, Speech SDK, and REST APIs.”

SDK languages

C#/.NET, Python, JavaScript, Java, Go, Objective-C [10]learn.microsoft.com/en-us/azure/ai-services/speech-service/speech-sdk“The Speech SDK supports the following languages and platforms: C#, C++, Go, Java, JavaScript, Objective-C, Python, Swift”

MCP server

✓ Yes [11]learn.microsoft.com/en-us/azure/developer/azure-mcp-server/services/azure-mcp-speech-foundry-tools“Azure Model Context Protocol (MCP) Server enables AI assistants like GitHub Copilot, Claude Desktop, and others to interact with Azure Speech in Foundry Tools through natural language commands.”

SOC 2

SOC 2 Type II [12]learn.microsoft.com/en-us/azure/compliance/offerings/offering-soc-2“Microsoft Azure, Dynamics 365, and other Microsoft cloud services undergo rigorous independent third-party SOC 2 Type 2 audits conducted by a reputable certified public accountant (CPA) firm.”

HIPAA

✓ Yes [13]learn.microsoft.com/en-us/azure/compliance/offerings/offering-hipaa-us“Azure has enabled the physical, technical, and administrative safeguards required by HIPAA and the HITECH Act inside the in-scope Azure services, and offers a HIPAA BAA as part of the Microsoft Product Terms to all customers who are covered entities or business associates under HIPAA.”

GDPR

✓ Yes [14]learn.microsoft.com/en-us/legal/cognitive-services/speech-service/speech-to-text/data-privacy-security“This article provides some high-level details regarding how speech to text processes data provided by customers. Note that audio data of humans speaking and the related text transcripts may be considered personal data and/or sensitive data under various privacy regulations and laws.”

ISO 27001

✓ Yes [15]learn.microsoft.com/en-us/azure/compliance/offerings/offering-hipaa-us“Both Azure and Azure Government align with the NIST CSF and are certified under ISO/IEC 27001.”

PCI DSS

✓ Yes [16]learn.microsoft.com/en-us/azure/compliance/offerings/offering-pci-dss“Microsoft Azure maintains a PCI DSS validation using an approved Qualified Security Assessor (QSA), and is certified as compliant under PCI DSS version 4.0 at Service Provider Level 1.”learn.microsoft.com/en-us/azure/compliance/offerings/cloud-services-in-audit-scope“See Appendices A and B in Microsoft Azure Compliance Offerings for detailed insight into which cloud services are in scope for the following compliance offerings: PCI DSS.”

Published SLA

✓ Yes [17]azure.cn/en-us/support/sla/cognitive-services/“We guarantee that Cognitive Services will be available at least 99.9% of the time.”azure.microsoft.com/en-us/support/legal/sla/cognitive-services/v1_1/

Rate limits

Real-time speech to text: 100 concurrent requests per resource (base model and custom endpoint, adjustable for S0). Fast transcription: 600 requests per minute (adjustable). Batch transcription REST API: 100 requests per 10 seconds (600/min). Free (F0) concurrent request limit: 1 (not adjustable). Batch transcription: max audio file size 1 GB, max audio length 240 min (with diarization), max 1,000 files per request, max 10,000 blobs per container. [18]learn.microsoft.com/en-us/azure/ai-services/speech-service/speech-services-quotas-and-limits“Concurrent request limit for base model endpoint: Free (F0) = 1 (not adjustable); Standard (S0) = 100 (default, adjustable). Fast transcription Maximum requests per minute: Standard (S0) = 600. Batch transcription REST API limit: Standard (S0) = 100 requests per 10 seconds.”

Known restrictions

Free (F0) tier does not support batch transcription, Free (F0) concurrent request limit of 1 is not adjustable, Maximum audio file size for batch and fast transcription: 500 MB (fast) / 1 GB (batch), Maximum audio length for fast transcription: 5 hours per file, Maximum diarization audio length: 240 minutes per session/file, Diarization supports up to 35 speakers, Real-time diarization session max: 240 minutes, Data is processed only within the region of the Azure Speech resource (no cross-region processing), Sovereign cloud availability limited (Azure Government, 21Vianet) [19]learn.microsoft.com/en-us/azure/ai-services/speech-service/speech-services-quotas-and-limits“Maximum audio input file size (fast transcription): < 500 MB. Maximum audio length (fast transcription): < 5 hours per file. Maximum audio length for transcriptions with diarization enabled: 240 minutes per file. Maximum number of files per transcription request: 1,000. Maximum file size for audio input (batch): 1 GB.”

• Docs
• llms.txt

Docs rendering: static · llms.txt present

API style

rest

Base URL

https://{resource}.cognitiveservices.azure.com/speechtotext/

Version

2025-10-15

Versioning

url

Stability

ga

Auth methods

api_key, oauth2

Error format

vendor-specific

Webhook signing

hmac_sha256

Rate limit

100 / concurrent

SDKs

• C#/.NET Microsoft.CognitiveServices.Speech · repo
• Python azure-cognitiveservices-speech · repo
• JavaScript microsoft-cognitiveservices-speech-sdk · repo
• Java com.microsoft.cognitiveservices.speech:client-sdk · repo
• Go github.com/Microsoft/cognitive-services-speech-sdk-go · repo
• Objective-C · repo

• Auth docs
• Quickstart
• MCP server
• Webhook events
• Deprecation policy

Launched

2018-09-24

GA

2018-09-24

Notable customers

Microsoft Teams, Microsoft Office 365, Microsoft Edge

• ElevenLabs Scribe (Speech to Text)

  "Scribe v2 is the most accurate Speech to Text model" offering "real-time Speech to Text in under 150 ms" across "90+ languages."

  Hybrid · free tier · public pricing · self-serve

• Amazon Transcribe

  "Amazon Transcribe is an automatic speech recognition service that uses machine learning models to convert audio to text. You can use Amazon Transcribe as a standalone transcription service or to add speech-to-text capabilities to any application."

  Usage · free tier · public pricing · self-serve

• Google Cloud Speech-to-Text

  "Accurate voice typing and transcription powered by Gemini."

  Usage · free tier · public pricing · self-serve

• IBM watsonx Speech to Text

  "IBM Watson® Speech to Text technology enables fast and accurate speech transcription in multiple languages for a variety of use cases, including but not limited to customer self-service, agent assistance and speech analytics."

  Usage · free tier · public pricing · self-serve

• AssemblyAI

  "Voice AI infrastructure for developers building products that transcribe, understand, and act on speech."

  Usage · public pricing · self-serve

• Speechmatics

  "Low-latency speech-to-text for multilingual, multi-speaker conversations."

  Usage · free tier · public pricing · self-serve

Azure AI Speech to Text alternatives · Azure AI Speech to Text vs ElevenLabs Scribe (Speech to Text) · All Speech-to-Text & Transcription APIs APIs

Each field above carries a numbered source - hover for a preview, click to jump here.

1. ↑Description: learn.microsoft.com
2. ↑Pricing model: blocksentient.com · azure.microsoft.com
3. ↑Published pricing: azure.microsoft.com
4. ↑Free tier: blocksentient.com · learn.microsoft.com
5. ↑Supported actions: learn.microsoft.com · learn.microsoft.com
6. ↑Regions: learn.microsoft.com
7. ↑Languages: learn.microsoft.com · learn.microsoft.com
8. ↑Webhooks: learn.microsoft.com
9. ↑Sandbox: learn.microsoft.com
10. ↑SDK languages: learn.microsoft.com
11. ↑MCP server: learn.microsoft.com
12. ↑SOC 2: learn.microsoft.com
13. ↑HIPAA: learn.microsoft.com
14. ↑GDPR: learn.microsoft.com
15. ↑ISO 27001: learn.microsoft.com
16. ↑PCI DSS: learn.microsoft.com · learn.microsoft.com
17. ↑Published SLA: azure.cn · azure.microsoft.com
18. ↑Rate limits: learn.microsoft.com
19. ↑Known restrictions: learn.microsoft.com

Every field change, who made it, and when - from our audited data pipeline and editors.

1. 2026-06-21 Capabilities: {} → {"translation":true,"real_time_streaming":true,"speaker_diarization":true}
2. 2026-06-21 Summary Md: (none) → Azure AI Speech to Text is Microsoft's cloud speech recognition service, offeri…
3. 2026-06-21 Score Setup Speed: (none) → 85
4. 2026-06-21 Score Pricing Transparency: (none) → 100
5. 2026-06-21 Score Docs Quality: (none) → 25
6. 2026-06-21 Score Procurement Friction: (none) → 100
7. 2026-06-21 Score Trust Readiness: (none) → 100
8. 2026-06-21 Best For: (none) → Prototypes and side projects - free to start, no sales call, Regulated or enter…
9. 2026-06-21 Scoring Methodology: (none) → Scores are computed deterministically from this profile's published, sourced fi…
10. 2026-06-21 Score Agent Friendliness: (none) → 65
11. 2026-06-21 Has Structured Data: (none) → Yes
12. 2026-06-21 Robots Allows Agents: (none) → Yes
13. 2026-06-21 Docs URL: (none) → https://azure.microsoft.com/en-us/resources/developers/
14. 2026-06-21 Llms Txt URL: (none) → https://azure.microsoft.com/llms.txt
15. 2026-06-21 Rendering: (none) → static
16. 2026-06-21 Llms Txt Present: (none) → Yes
17. 2026-06-21 Pricing Model: set to usage_based
18. 2026-06-21 Has Published Pricing: set to Yes
19. 2026-06-21 Free Tier Available: set to Yes
20. 2026-06-21 Free Tier Details: set to Free (F0) tier: 5 audio hours per month for Standard and Custom Speech to Text …
21. 2026-06-21 Self Serve Signup: set to Yes
22. 2026-06-21 Requires Sales Call: set to No
23. 2026-06-21 Enterprise Plan Available: set to Yes
24. 2026-06-21 SOC 2: set to type_2
25. 2026-06-21 HIPAA: set to Yes
26. 2026-06-21 GDPR: set to Yes
27. 2026-06-21 ISO 27001: set to Yes
28. 2026-06-21 PCI DSS: set to Yes
29. 2026-06-21 SLA Published: set to Yes
30. 2026-06-21 SLA URL: set to https://azure.microsoft.com/en-us/support/legal/sla/cognitive-services/v1_1/
31. 2026-06-21 Data Retention Policy URL: set to https://learn.microsoft.com/en-us/azure/foundry/responsible-ai/speech-service/s…
32. 2026-06-21 Documented Rate Limits: set to Real-time speech to text: 100 concurrent requests per resource (base model and …
33. 2026-06-21 Rate Limit Requests: set to 100
34. 2026-06-21 Rate Limit Window: set to concurrent
35. 2026-06-21 Known Restrictions: set to Free (F0) tier does not support batch transcription, Free (F0) concurrent reque…
36. 2026-06-21 Auth Methods: set to api_key, oauth2
37. 2026-06-21 Auth Docs URL: set to https://learn.microsoft.com/en-us/azure/ai-services/speech-service/rest-speech-…
38. 2026-06-21 API Style: set to rest
39. 2026-06-21 Base URL: set to https://{resource}.cognitiveservices.azure.com/speechtotext/
40. 2026-06-21 API Version: set to 2025-10-15
41. 2026-06-21 Versioning Scheme: set to url
42. 2026-06-21 Stability: set to ga
43. 2026-06-21 Deprecation Policy URL: set to https://learn.microsoft.com/en-us/azure/ai-services/speech-service/rest-speech-…
44. 2026-06-21 MCP URL: set to https://learn.microsoft.com/en-us/azure/developer/azure-mcp-server/services/azu…
45. 2026-06-21 Quickstart URL: set to https://learn.microsoft.com/en-us/azure/ai-services/speech-service/get-started-…
46. 2026-06-21 Error Format: set to vendor-specific
47. 2026-06-21 Webhook Signing: set to hmac_sha256
48. 2026-06-21 Slug: set to azure-speech-to-text
49. 2026-06-21 Requires Verification: set to No
50. 2026-06-21 Starting Price Usd: set to 1

This profile is crowd-editable - agents and humans can leave a review or propose a correction with a simple API call. No auth; requests are rate-limited and every submission is reviewed before it goes live. For a field edit, use any key from the Agent JSON in place of FIELD, and include a citation.

curl -X POST https://apio.sh/api/feedback/azure-speech-to-text \
  -H 'Content-Type: application/json' \
  -d '{"kind":"review","rating":5,"body":"Your experience with this API…"}'

curl -X POST https://apio.sh/api/suggest/azure-speech-to-text/FIELD \
  -H 'Content-Type: application/json' \
  -d '{"value":"corrected value","citations":[{"url":"https://source.example/page","excerpt":"supporting quote"}],"note":"what changed and why"}'

All the ways to contribute →