The WildcatTTS engine has been upgraded to Speech 2.8 Turbo (MiniMax via 302.ai), the latest generation of the MiniMax TTS model. This update also highlights interjection tags — a powerful feature for making your TTS messages sound more human.

Speech 2.8 Turbo

Speech 2.8 Turbo succeeds Speech 2.6 with continued improvements to synthesis quality, prosody, and latency. All supported voices have been upgraded automatically. The fallback to Speech-02 via Wavespeed AI remains in place for reliability.

No action is required — your voice, pitch, speed, emotion, and language settings all carry over.

Interjection Tags

You can embed interjection tags directly in your TTS messages to add realistic human sounds. The TTS engine will render these as natural vocalizations rather than reading them as text.

How to use them

Just include the tag in parentheses anywhere in your message:

!tts I can't believe that worked (laughs) that was amazing

!tts (sighs) here we go again

!tts Wait what? (gasps) No way!

Available tags

Tag Sound
(laughs) Laughter
(chuckle) Soft chuckle
(coughs) Coughing
(clear-throat) Throat clearing
(groans) Groaning
(breath) Breathing
(pant) Panting
(inhale) Inhaling
(exhale) Exhaling
(gasps) Gasping
(sniffs) Sniffing
(sighs) Sighing
(snorts) Snorting
(burps) Burping
(lip-smacking) Lip smacking
(humming) Humming
(hissing) Hissing
(emm) Hesitation / "umm"
(sneezes) Sneezing

These tags work with any voice and in any TTS mode. Mix them into your messages naturally for more expressive, lifelike delivery.

Full docs →