Gemini 3.1 Flash TTS, Text to Speech, Audio Output
Gemini 3.1 Flash TTS: lightning-fast, natural-sounding speech.
A tradução deste card ainda não está disponível; parte do conteúdo aparece em inglês.
Gemini 3.1 Flash TTS is a text-to-speech model that converts written text into high-quality, natural-sounding audio. It’s designed for speed and clarity, making it ideal for real-time applications like voiceovers, accessibility tools, and dynamic content creation.
Use this model when you need quick, reliable speech synthesis without robotic intonation. Its lightweight architecture delivers fast responses, so you can generate audio on the fly. It supports multiple languages and offers adjustable speed and pitch for customization.
What sets it apart is its balance of speed and expressiveness—it captures emotion and emphasis naturally, perfect for engaging listeners in podcasts, e-learning, or interactive experiences.
- Creating audiobooks from written content
- Adding voiceovers to videos or presentations
- Generating real-time narration for games
- Building voice-enabled accessibility features
- Producing dynamic announcements for events
Generates audio in milliseconds for near-instant playback.
Produces speech with human-like rhythm and intonation.
Supports multiple languages with accurate pronunciation.
- Use punctuation like commas and periods to control pacing.
- Add phonetic spelling in brackets for tricky names or acronyms.
- Adjust the speaking speed to match your project’s mood.
- Test short phrases first to fine-tune emphasis and tone.
- May struggle with very complex or unusual proper names.
- Longer texts may require splitting for optimal quality.
- Emotional range is good but not as nuanced as human speech.
Narrate in a warm, calm voice: 'Chapter One: The fog rolled in silently, shrouding the lighthouse in a blanket of white. Inside, the keeper stoked the fire, unaware of the storm brewing beyond the horizon.'
Energetic male voice: 'Welcome to The Creative Spark, where we explore the stories behind the world's most innovative ideas.'
Clear female voice, moderate speed: 'Today we'll learn how the water cycle works—from evaporation to condensation to precipitation. Follow along with the diagram.'
Fixes the generation to produce the same audio every time for a given text.
Shifts the voice higher or lower, useful for character voices.
Adjusts how fast the speech is delivered, from slow (0.5) to fast (2.0).
Limits the pool of possible intonation choices; lower values make speech more predictable.
Selects from available voice profiles, like 'en-US-Female-1' or 'en-GB-Male-2'.
Sets the spoken language, such as 'en' for English or 'es' for Spanish.
Controls randomness in pitch and tone; lower values sound more monotone, higher values more expressive.
Os preços exibidos são o que você paga na Muitto. Sem surpresas.
Can I customize the voice's accent or gender?
Yes, the model offers multiple voice profiles with different accents and genders to choose from.
How long does it take to generate one minute of speech?
Typically under a second, thanks to the Flash architecture.
Is the generated audio royalty-free for commercial use?
Yes, outputs can be used in commercial projects without additional licensing.