Google Unveils Gemini 2.5 Models with Native Text-to-Speech Capabilities
At Google I/O 2025, Google introduced significant updates to its Gemini 2.5 model series, notably integrating native text-to-speech (TTS) capabilities. This advancement positions Gemini as a formidable contender in the AI-generated speech domain, directly challenging offerings like OpenAI’s GPT-4o.
Native Text-to-Speech Integration
The Gemini 2.5 Pro and 2.5 Flash models now feature built-in TTS functionality. This integration allows developers to generate high-quality audio outputs directly from text inputs without relying on external services. The TTS system supports both single and multi-speaker outputs, enabling the creation of dynamic dialogues and narratives. Developers can fine-tune aspects such as voice style, accent, pace, and tone to suit specific application needs .
Multilingual and Expressive Speech Synthesis
Gemini’s TTS supports over 24 languages and can seamlessly switch between languages within a single audio stream. This feature is particularly beneficial for applications targeting diverse linguistic audiences. Additionally, the system can capture subtle vocal nuances, including whispers and emotional intonations, enhancing the realism and expressiveness of generated speech .
Enhanced Developer Tools and APIs
To facilitate the adoption of these new capabilities, Google has updated its Gemini API and Google AI Studio. Developers can now access:
- Asynchronous Function Calling: Ensures smooth user interactions by allowing the system to continue processing other tasks while executing functions in the background.
- Batch API: Enables the processing of multiple requests simultaneously, improving efficiency and reducing turnaround times .
These tools are designed to streamline the development process, allowing for rapid prototyping and deployment of applications leveraging Gemini’s advanced TTS features.
Availability and Future Outlook
The updated Gemini 2.5 models are currently available in preview through Google AI Studio and Vertex AI, with general availability expected in early June. As Google continues to refine these models, developers can anticipate further enhancements in speech quality, language support, and integration capabilities.
For more detailed information and to start building with Gemini’s new TTS features, visit the Google Developers Blog.
Below is an example.
Prompt.
[deep breath] ⚔️ ATTENTION, FORCES—FORM UP! ⚔️
[yelling] GOOGLE UNVEILS **GEMINI 2.5**—WITH NATIVE TEXT-TO-SPEECH FIREPOWER! [short pause]
(steady, commanding) At Google I/O 2025, we unleashed decisive upgrades to the Gemini 2.5 arsenal, integrating native TTS—placing Gemini in direct combat with OpenAI’s GPT-4o. [breath]
## (firm, rallying) Native Text-to-Speech Integration
(energetic) Gemini 2.5 Pro and 2.5 Flash now **speak for themselves**. No external gear needed. [whispering] Imagine code that turns into a living voice … instantly. [resume normal] Single-speaker, multi-speaker—choose your formation. Adjust voice style, accent, pace, and tone as precisely as you’d calibrate artillery. [short pause]
## (commanding, rising) Multilingual & Expressive Speech Synthesis
(steady) Over **24 languages**—and yes, seamless language-switching mid-sentence. [whispering] 英語から日本語へ、瞬時に… [back to authoritative] Capturing whispers, shouts, and every emotion between—so your dialogues **breathe**. [breath]
## (motivating) Enhanced Developer Tools & APIs
(yelling) NEW ORDERS: Asynchronous Function Calling—keep the operation moving while tasks execute undercover! Batch API—fire multiple requests in one volley, boosting efficiency and crushing turnaround times. [breath]
## (low, intense) Availability & Future Outlook
(steady) Gemini 2.5 is **in preview** on Google AI Studio and Vertex AI—full deployment expected early June. Stand ready for further boosts in speech quality, language coverage, and integration fire-support. [short pause]
[whispering] For mission docs and immediate enlistment, proceed to the Google Developers Blog. [breath-out] Dismissed.