Model positioning
A preview text-to-speech model in the Google Gemini 2.5 Pro series, designed for high-quality, controllable speech synthesis and suited to more complex speech generation prompts. (blog.google)
Capabilities and uses
- Accepts text and generates audio, supporting single-speaker speech and two-person dialogue. (aistudio.google.com)
- Speech style, intonation, emotion, speed, and pronunciation can be adjusted through natural-language prompts; supports multilingual speech generation. (blog.google)
- Can be used for announcements, story narration, podcasts, and voice acting for game characters, among other scenarios. (blog.google)
Limitations and version notes
- This model is a TTS preview model and should not be confused with the separate Live API model that supports real-time voice conversations. Current model documentation lists text input and audio output, with an input limit of 8,192 tokens and an output limit of 16,384 tokens; Live API, function calling, and context caching are not supported. These are the specifications for the undated API ID in the current documentation and do not necessarily represent all original specifications of the May 2025 version. (aistudio.google.com)
- The version identifier in this record is
gemini-2.5-pro-preview-tts-05-2025; Google's May 20, 2025 release notes use the API IDgemini-2.5-pro-preview-ttsand do not list the identifier with that date suffix as an invocable model ID. (aistudio.google.com) - On December 10, 2025, Google announced that upgraded TTS models would replace the models released in May; the current model page lists the most recent update as December 2025. Therefore, the specifications or prices for the current undated API ID should not be treated directly as guarantees for a separately invocable May 2025 snapshot. (blog.google)
Release date
May 20, 2025. (aistudio.google.com)
Pricing
Official Google Gemini Developer API pricing, lookup date: November 18, 2026; currency is US dollars (USD), and all prices are per 1 million tokens:
- Standard requests: text input $1.00; audio output $20.00.
- Batch processing: text input $0.50; audio output $10.00.
- The official pricing table marks the free tier as unavailable for both standard and Batch requests for this model. The prices above apply to the currently listed
gemini-2.5-pro-preview-ttsAPI ID; separate historical pricing for the May 2025 version has not been verified. (aistudio.google.com)