ClosedGoogle/10 trustVerified

Gemini 2.5 Pro TTS

Name: Gemini 2.5 Pro TTS
Rating: 8

Google's quality-first TTS: steady long-form narration with style direction and two-speaker dialogue.

About

Gemini 2.5 Pro TTS is the quality-first sibling in Google's 2.5 speech lineup, aimed at long-form and structured audio rather than real-time chat. It shares the same 30-voice roster and 70-plus languages as the Flash model, but is tuned to hold pacing and tone steady across a whole chapter or episode, and it takes natural-language style direction ('read this as a calm documentary narrator') plus two-speaker dialogue in a single request. Output comes in six formats including MP3, WAV, Opus, AAC and FLAC.

Like the Flash model it shipped in preview in May 2025 and received the December 2025 expressivity update. Pricing is roughly $1 per million text input tokens and $20 per million audio output tokens — confirm the current numbers on Google's pricing page, since they have moved before. No independent benchmarks exist; the quality case rests on Google's own reporting and the December benchmark claim it shares with the Flash model.

Specs

Context window

TTS request/response — text in, long-form audio out; 24kHz mono, six output formats

Pricing

Token-based: ~$1/1M text input tokens, $20/1M audio output tokens (confirm on Google's pricing page)

Free quota

No free tier on Pro — paid tier only

API access

API available — gemini-2.5-pro-preview-tts via the Gemini API

Weights

Closed — API only

Released

Not disclosed

Benchmarks

No independent published benchmarks; quality positioning is provider-reported

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

Use cases

text-to-speechaudiobookspodcastingmultilingual-ttse-learning

Trust score — 8/10

/10 trustVerified

Scored 8/10 (TIER verified, metrics-v1). Evidence: Google Developers Blog (Dec 10, 2025) covering both Flash and Pro updates; public speech-generation docs; model live in AI Studio. Strengths: best-in-lineup quality for long-form, two-speaker dialogue in one request, style prompting, six output formats. Docked for: (1) no independent benchmarks; (2) paid-only with pricing that has shifted (verify before budgeting); (3) preview status. License: proprietary (Google service).

How the trust score is built

Full methodology →

Adoption · stars/downloads, log-scaled vs peers: 7.0/10
Maintenance · recency, releases, issue load: 8.0/10
Openness · license + docs: 3.0/10
Manual review · sanity read + red-flag check: 8.0/10

· Quality-first TTS, preview May 2025; same 30 voices / 70+ languages as Flash
· Two-speaker dialogue in one request; natural-language style direction; six output formats
· Pricing ~$1/1M text in, $20/1M audio out — verify on pricing page; paid tier only
· Google logo staged