# Google launches Gemini 3.8 Flash and Flash‑Lite text‑to‑speech models

Digest AI · Generative AI & Models · published 2026-09-23T15:25:14Z · updated 2026-09-23T15:50:02Z

Canonical: https://digestai.news/story/google-launches-gemini-3-8-flash-and-flashlite-texttospeech-models

## Summary

Google added two text‑to‑speech models to its Gemini family: Gemini 3.8 Flash TTS, aimed at deep creative direction and character design, and Gemini 3.8 Flash‑Lite TTS, built for high‑volume, cost‑efficient scaling. Both models let developers and enterprises generate expressive, multilingual speech using natural‑language prompts, with control over pacing, dialect, back‑channeling and line‑by‑line performance.

The Flash model can create new voices from scratch, replicate a voice from a 30‑second sample, and offers an expanding library that already includes more than 2,000 production‑ready voices across over 100 languages and dialects, such as Mexican Spanish and Quebec French. The Flash‑Lite variant focuses on dubbing and large‑scale audio creation. In benchmark tests, Gemini 3.8 Flash TTS ranked #1 on Hume AI’s Voice Design Benchmark with a score of 71.4 and led accent modeling with 60.8, while Flash‑Lite took the #2 spot on the Overall Quality Index. The models are available today through the Gemini API, Google AI Studio, Gemini Notebook and, soon, Gemini Enterprise.

## Key points

- Gemini 3.8 Flash TTS creates new voices from prompts and replicates from a 30‑second sample
- Library includes 2,000+ production‑ready voices in more than 100 languages and dialects
- Flash TTS ranked #1 on Hume AI’s Voice Design Benchmark (71.4) and leads accent modeling (60.8)

## Why it matters

The models give developers a scalable way to produce high‑quality, multilingual speech, opening new possibilities for games, podcasts, dubbing and interactive voice agents.

## Sources

1. [Gemini 3.8 text-to-speech says hello](https://deepmind.google/blog/say-hello-to-gemini-38-text-to-speech) (Google DeepMind, 2026-09-23, primary source)
2. [Google releases Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, its "most expressive audio generation models yet", with support for more than 100 languages](https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-8-text-to-speech) (blog.google, 2026-09-23, primary source)
3. [Google’s new Gemini TTS models can clone a voice from 30 seconds of audio](https://thenextweb.com/news/gemini-tts-3-8-flash-voice-design-cloning) (thenextweb.com, 2026-09-23)
4. [Google Rolls Out Gemini 3.8 Speech Models In API And AI Studio](https://unite.ai/google-rolls-out-gemini-3-8-speech-models-in-api-and-ai-studio) (Unite.AI, 2026-09-22)

Part of the developing story: [Google Gemini Voice Model Saga](https://digestai.news/thread/google-deepmind-launches-gemini-3-8-live-and-3-8-live-extended-thinking-models) (3 stories)

## Cite

Digest AI, "Google launches Gemini 3.8 Flash and Flash‑Lite text‑to‑speech models", 23 September 2026, https://digestai.news/story/google-launches-gemini-3-8-flash-and-flashlite-texttospeech-models

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/google-launches-gemini-3-8-flash-and-flashlite-texttospeech-models.json