Google Releases Gemini 3.8 Flash TTS and Flash-Lite TTS, Opens Omni Video Generation in Vids

The TTS release splits the models into two tiers. Gemini 3.8 Flash TTS is built for deep creative direction and character design, with target uses including gaming, immersive audiobooks, podcasts and interactive media. It offers granular control over acting cues, pacing, dialect shifts and backchanneling. Gemini 3.8 Flash-Lite TTS is built for high-volume, cost-efficient scale, with Google positioning it for dubbing, audio content creation and expressive voice agents. It offers fine-grained control over tone, pacing and expressive nuance. Both models let developers direct delivery line by line using natural language. They are rolling out through the Gemini API and Google AI Studio. Access is API-only, with no open weights for self-hosting. Enterprise API access via Gemini Enterprise is listed as coming soon.

MarkTechPost reported that previous Gemini TTS offered 30 original voices. The 3.8 release moves to a much larger voice system. Flash TTS can create new voices from prompts describing role, accent and voice characteristics, a feature that works across more than 100 languages and dialects. Google's demos include a Melbourne DJ, a monotone robot and a Japanese dragon. Developers get more than 2,000 production-ready voices, covering regional varieties such as Mexican Spanish, Quebec French and Scots English. Custom voices can be saved and reused with minimal drift across projects. Voice remixing, listed as coming soon, will let users adjust a library voice's timbre, pitch, pace and accent through prompts.

Both models accept stage directions written in the script, and Gemini can also steer delivery from natural script cues, according to MarkTechPost. The report said voice quality, pacing and timbre hold across hours of continuous audio. A single script can drive a multi-turn conversation with distinct, separated voices through native two-speaker staging. Vocal bursts, or non-verbal cues, add conversational texture, while backchanneling interjections such as |mhm| and |yeah| control reaction beats and comedic timing.

Voice replication builds a consistent vocal profile from a 30-second audio sample. MarkTechPost reported the sample must be the user's voice or one they have rights to use, and replication requires a verbal consent recording from the voice owner matched against the reference speaker. Every clip from Gemini Audio models carries a SynthID watermark embedded directly in the audio output. Replicated voices also carry C2PA content credentials. Google points to the Gemini 3.8 Audio model card for its broader safety approach. The report added that voice replication is unavailable in AI Studio in several regions, including the UK, EEA and India.

Google reported benchmark results for the new TTS models, according to MarkTechPost. On the Hume AI Voice Design Benchmark, Flash TTS ranks first overall with a score of 71.4. In accent modeling, Flash TTS leads with a score of 60.8. On the Hume AI Overall Quality Index, Flash TTS ranks first and Flash-Lite TTS ranks second. In Voice Arena blind preference, both models take top positions in Japanese, Brazilian Portuguese, Vietnamese, Modern Standard Arabic, Mexican Spanish and Hindi.

Android Authority reported that Google Vids is expanding access to anyone with a personal or Workspace account. Users will be able to generate 1080p videos with Gemini Omni 1.1 Flash at no cost. Google will soon roll out Gemini 3.8 Flash-Lite text-to-speech in Vids for voiceovers. The report said Vids launched in 2024 as a tool for professionals. Even as access expanded beyond Gemini for Workspace customers, the AI-powered video generator remained behind Google's subscription plans. Google gave free users a taste earlier this year, and the latest change opens the door more broadly. Android Authority referred to both Gemini Omni 1.1 Flash and Gemini Omni 1.1 Flash-Lite in describing the Vids integration; its summary said users can generate 1080p videos with Gemini Omni 1.1 Flash at no cost, while another sentence named Gemini Omni 1.1 Flash-Lite as one of the integrated models.