Post-Cutoff
Search For AI agents

Google DeepMindGemini 3 familyCurrentSpeech and audioReleased 22 Sep 2026

Gemini 3.8 Flash TTS

Studio-grade, steerable text-to-speech (audiobooks, games, dubbing, voice agents).

Model idgemini-3.8-flash-tts

Price$0.50 in, $9 out per 1M tokens

Checked29 September 2026

Where it stands

19newer speech and audio models since its release

The 5 newest are listed. All speech and audio models

Before it from Google
Gemini 3.8 Live15 Sep 2026
Weights
Closed weights; access is hosted only.

How to call it

2 ways to use Gemini 3.8 Flash TTS, with the model id each provider expects.

Providers, model ids, endpoints and docs for Gemini 3.8 Flash TTS
ProviderModel idEndpointDocs
Gemini API gemini-3.8-flash-tts https://generativelanguage.googleapis.com/v1beta/models/gemini-3.8-flash-tts:generateContent ai.google.dev
Gemini API gemini-3.8-flash-lite-tts ai.google.dev

Pricing

Prices for Gemini 3.8 Flash TTS
Input$0.50 per 1M tokens
Output$9 per 1M tokens

As published: “per 1M tokens (text in / audio out; introductory through 2026-12-31, $1.00 / $18.00 from 2027-01-01)”.

Source: ai.google.dev, checked 29 September 2026.

Specs

Input
Text
Output
Audio
Context window
8,192 tokens
Max output
16,384 tokens
Open weights
No
Licence
Proprietary
Released
22 September 2026
Status
Current
Type
Speech and audio

What stands out

What its maker marketed at launch, and what people found later, each with its source.

  • Voice design from prompts

    Create entirely new voices from natural-language descriptions; #1 on Hume AI Voice Design Benchmark.

    Source: blog.google

  • #1 on Hume Real-World VoiceEQ leaderboard

    Found after launch

    Hume’s blind human-rated benchmark (2026-09-24): Gemini 3.8 Flash TTS 0.920 and Flash-Lite TTS 0.914 expressivity-reliability score, ahead of Gemini 2.5 Pro TTS (0.880) and Cartesia Sonic 3.6 (0.840); long-form stability up from 1.22 to ~2.9-3.0/5, but weaker speaker similarity (3.68/5).

    Source: hume.ai

  • Voice replication

    Recreates a consistent voice from a ~30-second sample with consent verification; 2,000+ library voices.

    Source: blog.google

  • Directed long-form multi-speaker audio

    Line-by-line direction of pacing/emotion, dual-speaker staging, stable over hours; 130+ languages with auto-detection.

    Source: ai.google.dev

Notes

Sibling gemini-3.8-flash-lite-tts (101 languages) costs $0.50 in / $6.00 audio out (intro); own file gemini-3-8-flash-lite-tts. Outputs SynthID-watermarked. Older: gemini-3.1-flash-tts-preview, gemini-2.5-flash-preview-tts, gemini-2.5-pro-preview-tts.

curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-3.8-flash-tts:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" -H "Content-Type: application/json" \
  -d '{"contents":[{"parts":[{"text":"Say cheerfully: Have a wonderful day!"}]}],
       "generationConfig":{"responseModalities":["AUDIO"]}}'

Sources: model page, pricing, launch blog.