Skip to navigation
Text-to-speech

Emotion styles

What the speechify:style emotion tag does on each model

Emotion styles are applied only by the retired Simba 1.6 models, simba-english and simba-multilingual. The current models, simba-3.2 and simba-3.0, accept <speechify:style emotion="..."> and return 200, but the style has no effect on the audio. New workspaces cannot select Simba 1.6, and from 2026-11-21 both Simba 1.6 ids are served by the current models. See Migrating off Simba 1.6.

On the current models

simba-3.2 and simba-3.0 apply three SSML tags: <break> for pauses, <prosody rate> for speaking rate, and <sub alias> for pronunciation. They accept <prosody pitch>, <prosody volume>, <emphasis> and <speechify:style> without applying them. See SSML for each tag.

The tag on Simba 1.6

On Simba 1.6, wrap a span in <speechify:style> and set emotion:

<speak>
  <speechify:style emotion="cheerful">
    Great news! Your order has been shipped!
  </speechify:style>
</speak>

Simba 1.6 values: angry, cheerful, sad, terrified, relaxed, fearful, surprised, calm, assertive, energetic, warm, direct, bright. An unrecognised value is ignored, not refused.