Skip to main content
Audio tags add vocal actions and non-speech sounds at specific points in text-to-speech input. Breeze supports the same 34 audio events across all 23 languages in the current public language contract. Use the exact tag spelling for the language being spoken. English tags use parentheses; every other supported language uses square brackets. Do not translate a tag yourself or mix tag syntaxes within a request.
Audio-tag support does not override model language support. Call GET /v1/models and inspect each model’s languages array before choosing a model-language pair.

Tag reference

The canonical event names on the left identify the shared audio action. Send the exact language-specific tag shown on the right as part of your input text.

Arabic (ar)

Czech (cs)

German (de)

Greek (el)

English (en)

Spanish (es)

Finnish (fi)

French (fr)

Hindi (hi)

Indonesian (id)

Italian (it)

Japanese (ja)

Korean (ko)

Dutch (nl)

Polish (pl)

Portuguese (pt)

Romanian (ro)

Russian (ru)

Thai (th)

Turkish (tr)

Ukrainian (uk)

Vietnamese (vi)

Simplified Chinese (zh)