MAI-Voice

Voice & speech playground

Hydrating page from local storage...

Create expressive speech with locale-specific voices and SSML style controls. Text is synthesized, not sent to a chat model.

Microphone access requires HTTPS or localhost and your permission. Audio is sent to the selected provider only when you submit a request or start a live session. Use headphones for live conversations.

Find models in Microsoft Foundry. MAI speech models use Azure Speech APIs and a supported Speech resource region, not a chat deployment endpoint.

Connection

Use a Speech-capable resource in a supported region.

Endpoint

Authentication Mode

API Key

Advanced

Speech controls

Language / locale

Choose a documented language or locale, or select Custom to enter another provider-supported code. Match the selected MAI voice's locale and provide text in that language.

Voice

Choose a built-in voice, or select Custom to enter another documented voice ID. Grok also accepts an existing authorized custom voice ID.

Output format

24 kHz mono audio with native playback and download.

Expressive style (optional)

Supported styles vary by voice. The API documentation lists each voice's styles.

Style strength

Dictation resource endpoint (optional)

Uses the same Speech resource key. Required for dictation when synthesis uses a regional TTS endpoint.

Ready. Configure your connection to begin.

Optional microphone dictation

Audio uploads are limited to 25 MB in this playground; microphone recordings stop at five minutes. Dictation makes a separate, billable MAI-Transcribe-2 call and replaces the text below only after it succeeds.

Text to speak

Provider usage is billable. No estimated price is assumed; returned usage is available in Payloads.

Generated speech will appear here with playback and download controls.