Microsoft's speech APIs for speech-to-text, text-to-speech, translation, and speaker recognition, now part of Foundry Tools.
Azure Speech in Foundry Tools — product homepage
1 / 2Overview
What is Azure AI Speech?
Add voice to your apps and AI agents with Microsoft's speech AI.
Azure AI Speech (now Azure Speech in Foundry Tools) provides prebuilt, customizable, and multilingual speech AI models as managed APIs: speech to text, text to speech, translation, and speaker recognition.
- Transcribe calls and meetings in real time
- Generate natural speech, including a custom neural voice for your brand
- Translate speech and text across many languages
- Build voice-enabled AI agents on the same stack
Billing is pay-as-you-go with no upfront cost: speech-to-text and translation are billed per audio hour, text-to-speech per character, and speaker recognition per transaction. You can try Azure free for up to 30 days.
Platforms and languages
Azure AI Speech Availability
Platforms
Languages
Capabilities
Azure AI Speech Key Features
Speech to text
Fast transcription for call-center calls and meetings, with audio captions and global transcription across 100+ languages.
Text to speech
Natural-sounding prebuilt voices, plus custom neural voice to build a brand-specific voice and speaking style.
Speech and text translation
Real-time translation across a growing list of languages, customizable for specific industries.
Voice for AI agents
Combine foundation models with custom audio input and output models to build AI agents that speak.
Embedded speech
On-device speech-to-text and text-to-speech for situations with unstable or no cloud connection.
Best for
Who uses Azure AI Speech?
Developers adding voice to apps and AI agents
Add speech-to-text, text-to-speech, and translation to applications via managed APIs
Plans and access
Azure AI Speech Pricing
Paid
Common questions
Azure AI Speech FAQs
What is Azure AI Speech?
Azure AI Speech is Microsoft's set of managed speech APIs — speech to text, text to speech, translation, and speaker recognition — now grouped under Foundry Tools as Azure Speech.
How is Azure AI Speech priced?
It is pay-as-you-go with no upfront cost: transcription and translation are billed per audio hour, text-to-speech per character, and speaker recognition per transaction. New users can try Azure free for up to 30 days.
Can Azure AI Speech create a custom brand voice?
Yes — custom neural voice lets you build a branded voice and speaking style on top of the prebuilt voices.
Does Azure AI Speech work offline?
Yes, embedded speech supports on-device speech-to-text and text-to-speech when cloud connectivity is unstable or unavailable.
Reviews
See what the community thinks and share your experience.
—
Based on 0 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
