Google launched a voice upgrade to its model – and managed to stir up anger online

Google has launched its Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS speech generation models, enabling customized voice design and complex dialogue direction. The models achieve record scores in global quality tests, but the launch sparked online outrage due to the delay of the Gemini 4 Pro flagship model, which developers have been waiting months for.

Google announced over the weekend the launch of two text-to-speech models: Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS. The new models allow developers to create customized voice identities in over 100 languages and dialects from a verbal description alone, and offer a line-by-line script direction mechanism including whispers, pace changes, accents, and multi-speaker conversations. In global quality tests, the model took first place in the Hume AI index for voice design and audio quality, and led in blind evaluations on Voice Arena. However, the launch met with a wave of scathing criticism from the AI community, which has lost patience with yet another "Flash" version and has been waiting months for the Gemini 4 Pro flagship model. Social networks X and Reddit were flooded with memes and mockery of Google's release pace, while its competitors (OpenAI, Anthropic, Elon Musk) have launched massive frontier models. Developers expressed frustration that Google continues to split its brand into minor versions instead of releasing the flagship model. The models were integrated starting today into Google AI Studio, Google Vids, Gemini Notebook, and partnerships with Figma and HeyGen.