ClosedGoogle/10 trustVerified

Gemini 2.5 Flash TTS

Name: Gemini 2.5 Flash TTS
Rating: 8

Google's low-latency native TTS: 30 prebuilt voices, 70+ languages, style-prompted speech from the Gemini API.

Visit official page →

About

Gemini 2.5 Flash TTS is Google's low-latency text-to-speech model, released in preview in May 2025 and updated that December with stronger expressivity, context-aware pacing and more consistent multi-speaker dialogue. It speaks through 30 prebuilt voices with star-derived names (Kore, Charon, Zephyr and the rest), each able to narrate in more than 70 languages, and returns 24kHz mono audio. Developers reach it through the Gemini API by requesting AUDIO output from gemini-2.5-flash-preview-tts, or can experiment in AI Studio and the Gemini app.

Pricing is token-based: $0.50 per million text input tokens and $10 per million audio output tokens, with Google noting that one second of audio equals 25 audio tokens. There is a free tier in AI Studio, though Google says free-tier data is used to improve its products. Google claims the updated models lead in most categories of Hume AI's TTS benchmark — a provider claim with no independent verification I could find.

Specs

Context window

TTS request/response — text in, up to ~10 minutes of audio per request; 24kHz mono PCM output

Pricing

Token-based: $0.50/1M text input tokens, $10/1M audio output tokens (25 tokens = 1 second of audio)

Free quota

Free tier available in AI Studio (free-tier data used to improve Google's products, per Google)

API access

API available — Gemini API generateContent with AUDIO response modality (gemini-2.5-flash-preview-tts)

Weights

Closed — API only

Released

Not disclosed

Benchmarks

Google claims the Dec 2025 update leads in most categories of Hume AI's TTS benchmark — provider-claimed, no independent verification

Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.

Use cases

text-to-speechvoice-agentsmultilingual-ttspodcastingaudiobooks

Trust score — 8/10

/10 trustVerified

Scored 8/10 (TIER verified, metrics-v1). Evidence: Google Developers Blog (Dec 10, 2025) documenting the May 2025 release and December update; public speech-generation docs; token pricing on Google's pricing page. Strengths: strong multilingual coverage, transparent token pricing, free tier, style prompting. Docked for: (1) benchmark leadership claim is provider-reported with no independent verification; (2) preview status — APIs may change before GA; (3) free-tier data reuse policy. License: proprietary (Google service).

How the trust score is built

Full methodology →

Adoption · stars/downloads, log-scaled vs peers: 8.0/10
Maintenance · recency, releases, issue load: 8.0/10
Openness · license + docs: 3.0/10
Manual review · sanity read + red-flag check: 8.0/10

· Released in preview May 2025; Dec 2025 update added expressivity, context-aware pacing, multi-speaker consistency
· 30 prebuilt voices, 70+ languages, 24kHz mono via Gemini API (AUDIO modality)
· Pricing: $0.50/1M text in, $10/1M audio out (25 tokens = 1s audio); free tier in AI Studio
· Google logo staged