Google launches Gemini 3.8 Flash TTS and Flash-Lite TTS

Google announced two new text-to-speech models on September 23, 2026: Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, which the company calls its most expressive audio generation models yet.

Both are available through Google AI Studio, the Gemini API, Gemini Enterprise, Gemini Notebook and Google Vids, continuing the Gemini Audio line.

Two models, two jobs

Gemini 3.8 Flash TTS is built for creative direction and character design: new voices are created from scratch with natural language prompts, with line-by-line control over acting cues, pacing, dialect shifts and backchanneling — for gaming, audiobooks, podcasts and interactive media. Flash-Lite TTS targets high-volume, cost-efficient work such as dubbing and expressive voice agents, with fine-grained control over tone, pacing and nuance.

Custom voices and voice replication

Voice design scales from 30 original voices to an unlimited library: prompts set role, accent and other vocal characteristics across more than 100 languages and dialects, and a catalogue of 2,000+ ready voices adds regional varieties from Mexican Spanish to Scots English. Voice replication rebuilds a vocal profile from a 30-second sample, protected by consent verification, SynthID watermarking and C2PA credentials; AI Studio replication is unavailable in Illinois, Texas, the EEA, the UK, Switzerland and India.

Direction, long-form and benchmarks

Both models let creators steer delivery with script cues or written stage directions. Google highlights long-form generation that holds timbre and pacing across hours of audio with minimal speaker drift, two-speaker staging from a single script, and scripted vocal bursts such as <laughs> and <sigh>. On Hume AI's Voice Design Benchmark, Gemini 3.8 Flash TTS took the top overall score (71.4) and led accent modeling (60.8); Hume's Overall Quality Index ranked it first and Flash-Lite second. Blind preference tests on Voice Arena placed the models among the leaders in Japanese, Brazilian Portuguese and Hindi.

Availability

Flash TTS is available now to developers in the Gemini API and Google AI Studio and to everyone in Gemini Notebook, with API access in Gemini Enterprise coming soon. Flash-Lite TTS is live in the Gemini API, AI Studio and Google Vids. A new AI Studio audio playground adds a voice design workspace and a dual-speaker screenplay editor, and Agora, LiveKit, Pipecat and Vercel are integrating the models.