Addy Crezee
![Woman speaks into vintage mic surrounded by neon [whisper], [panic], [excited] tags — Gemini 3.1 Flash TTS emotion control.](/content/images/size/w1200/2026/05/gemini-3-1-flash-tts-emotion-tags-article-cover.jpg)
google just dropped gemini 3.1 flash tts
you can now control tone, pace, and emotion with simple text commands inside the prompt like: [excited], [whisper], [panic] – just by inserting them into the text
btw it’s now the #2 voice model on the artificial analysis tts leaderboard
we just unlocked:
• ai podcasts that don’t sound robotic
• game characters that act, not just speak
• customer support with actual personality
• localized content that truly feels native
Gemini 3.1 Flash TTS is our most controllable text-to-speech model yet.
— Google DeepMind (@GoogleDeepMind) April 15, 2026
With new Audio Tags, you can easily direct vocal style, delivery, and pace through text commands. 🧵 pic.twitter.com/Bq4SD8eLUN
just check this out! crazy
0:00
/0:33
