The new Gemini 3.1 Flash TTS text-to-speech model from Google offers developers and enterprises a high degree of control and expressiveness, opening up new possibilities for creating advanced AI applications.
Google has introduced the new Gemini 3.1 Flash TTS text-to-speech model, available on Google AI Studio and Vertex AI. This model provides developers and enterprises with a high level of manageability and expressiveness, enabling the creation of advanced AI speech applications. The new TTS product supports over 70 languages and ensures clear and high-quality sound.
Gemini 3.1 Flash TTS offers over 200 audio tags that allow control over the pace, expressiveness, and style of playback. Users can choose from 30 base voices and customize the target language, forming the foundation of the audio output. Tags such as [whispers] and [happy] can be entered directly into the text, simplifying the management of pace and intonation.
This technology allows programming speech style using simple text commands, such as controlling accent or tone. The ability to combine commands with texts in different languages, as demonstrated with French, opens new horizons for multilingual applications and facilitates customization to specific requirements.
TTS solutions are key in creating accessible digital spaces. Gemini 3.1 Flash TTS will impact various fields, including game soundtracks and audiobooks, making them more accessible to a wider audience. This is especially important for people with disabilities and those who prefer audio content in their daily lives.
The implementation of the new model also includes the use of SynthID watermarking, which is embedded in the audio output and helps identify AI-generated content, combating counterfeiting and copyright violations.
Sources
Replies (0)
No replies in this topic yet.
