Whispering to models at Google DeepMind
๐ชด London
- Tweets 13,289
- Following 1,722
- Followers 83,875
- Likes 36,029
Gemini 3.8 Flash TTS is out, and it can sing (a bit) too. It works out the tune by itself.
Style: Singing
Transcript:
Happy release day to you,
Happy release day to you,
Happy release day, dear Logan,
Happy release day to you.
Introducing Gemini 3.8 Flash and Flash-Lite TTS, our new SOTA text to speech model with:
- a new voice design experience
- 2,000+ production ready voices
- voice replication
- support for 100 languages
- voice remixing (soon)
- #1 spot on Hume AI's voice benchmarks
and more!!
23
14
314
28,131
Voice design is fun:
> A 1930s Mid-Atlantic newsreel commentator. Brassy, forward-placed theatrical tenor with crisp trilled r's, punchy staccato phrasing, and vintage ribbon-microphone midrange projection.
Create and deploy custom audio with our new text-to-speech models:
๐ต Gemini 3.8 Flash TTS: Design unique voices with distinct accents and characteristics.
๐ต Gemini 3.8 Flash-Lite TTS: Built for efficiency and scale, choose from your created styles or our expansive production-ready library.
4
3
102
12,393
Come and work on Omni at Google DeepMind
We are hiring research scientists and engineers๐งโ๐ฌ on the Gemini Omni team! Both in the US (Mountain View / SF) and Europe (London / Zรผrich).
Apply here: google.com/about/careers/appโฆ
2
37
4,218
ShadowBat ๐ฑ๐ง
Come and work on world models in London at DeepMind ๐
We are hiring a Research Engineer in London to work on real-time generative world models. This is an exceptional opportunity to advance and scale cutting-edge diffusion models and multimodal LLMs: google.com/about/careers/appโฆ
3
2
43
6,296
