Give your ideas a voice.
And a soundtrack.
Turn words into speech, create with a reference voice, compose music or shape a sound effect. Bring the sound of your next story into the same Studio as your images and videos.
Create audio in Studio Speech · Speech from reference · Music · Sound effectsFind the voice for your story.
Listen to a published voice sample, then explore more voices in Studio.
- Voice
- Heart
- Language
- English
- Accent
- American
Choose the sound that fits your story.
Explore speech, reference voices, music and sound effects in one audio workspace. Choose from available audio models, with controls that adapt to the selected model. Voice, language, delivery style, speed, lyrics and duration options depend on the model you choose.
From written direction to something you can hear.
Start with the kind of audio you need. Studio keeps your text, model settings and result together.
Choose your starting point
Use Speech for text-to-speech, Speech from reference to guide a voice with an uploaded clip, Music for a soundtrack, or Sound effects for a sound or atmosphere. Studio shows compatible models for each mode.
Direct the voice or sound
For speech, write the words you want spoken and choose a supported voice and language. Listen to preset voices before choosing one. Reference speech accepts an MP3 or WAV clip, with limits shown for the selected model. For music and effects, describe the sound and use the available duration, lyrics, instrumental or looping controls.
Review the price, then listen
Check the exact credit price before creating. Your finished audio appears in the conversation with playback, seeking, speed and download controls. Keep the result in your Library or download it in the model’s output format.
Four ways to bring sound into the picture.
- Speech: narration, dialogue, voiceovers and spoken ideas
- Speech from reference: new words guided by an uploaded voice clip
- Music: soundtracks, musical sketches and supported lyric-led songs
- Sound effects: atmosphere, textures, transitions and individual sounds
- Voice previews, language and delivery controls on supported models
- A consistent player with seeking, playback speed and audio downloads
Before your first audio creation.
Can I upload a reference voice?
Yes. Choose Speech from reference and upload an MP3 or WAV clip. Studio filters the model list to compatible reference-voice models and shows the selected model’s requirements. Use a voice you own or have permission to use. A reference guides a new generation; it does not guarantee an identical voice.
Can I listen to voices before choosing one?
Yes. Play the selected voice from Audio settings, or browse the model’s preset voices and listen before choosing one. Previews do not use your credits. The first preview of a voice may take a moment to prepare.
How do I choose the language?
Supported speech models show a Language setting, including automatic detection where available. Write your text in the language you want spoken. Choosing a language guides pronunciation; it does not translate your text. Languages and voice options vary by model.
Can I control the delivery, lyrics or length?
The controls follow the selected model’s capabilities. Supported speech models offer speed, style instructions or voice tuning. Music models may offer lyrics and instrumental output, while supported music and effects models expose duration or looping controls.
Which audio models can I use?
The Audio model picker includes available audio models, filtered for Speech, Speech from reference, Music or Sound effects. Check the current list in Studio for model names, supported inputs and settings.
How is audio priced and downloaded?
The selected model, text or duration determines the price. Studio shows the exact credit amount before creation. Download the finished audio from its player; MP3, WAV and other supported output formats depend on the model.
Your next creation has a sound.
Start with a few words, a reference voice or a sound you can already imagine.