Learn how to write effective text prompts to generate sound effects using the Text to sound effects (beta) and Voice to sound effects (beta) features.
The Text to sound effects (beta) and Voice to sound effects (beta) features let you generate sound effect audio clips using multiple input methods— written prompts and recorded audio hints. By combining a text prompt with a voice performance, you can generate a sound effect that matches the description of your text prompt while following the timing and energy of your voice performance. Whether you're crafting atmospheric soundscapes, impactful effects, or ambient background noise to enhance your visual storytelling, a well-written text prompt is key to unlocking the full potential of these features.
Provide clear, concise, and direct text descriptions of the sounds you want to generate. This approach emphasizes the auditory characteristics and focuses on the sound itself—what it actually sounds like.
Examples of Do's
- lion roaring
- heavy rain on a metal roof
- crackling campfire
Examples of Don'ts
- the sound of a lion roaring
- the pitter-patter sound made by raindrops falling onto a corrugated iron rooftop
- the sound of wood burning in a fire pit with occasional pops and hisses
In your text prompts, include clear adjectives to describe the qualities of the sound effects and verbs to convey the action or behavior of the sound. This guides the detailed characteristics of the sound.
Examples of Do's
- Very loud explosion or Soft explosion
- Forceful ocean waves crashing on the shore or ocean waves gently lapping on the shore
- Porcelain cup dragged over a wooden table
You can use comma-separated keywords or descriptions to quickly specify multiple characteristics of the desired sound. This approach also helps you adhere to other best practices in prompt writing, while still crafting concise text descriptions that effectively convey the complex sound effects you want to generate.
Examples of Do's
- Robot, scifi, futuristic
- Cinematic impact, sharp attack
- Orchestra hit, low pitch, dramatic trailer
Generate Sound Effects is designed to generate one sound at a time, to ensure maximum quality and control. To generate soundscapes that combine multiple sounds, generate each sound separately and layer them using multiple audio tracks. You can also leverage the post-generation edit options that let you fine-tune timing and volume, and download each sound for editing or use in your projects.
Examples of Do's
- footsteps on snow
- desert wind
- bird calls
Examples of Don'ts
- footsteps on snow with desert wind in the background followed by bird calls
When generating ambient soundscapes, it's often more effective to use broad or general descriptions rather than highly specific ones. Overly detailed prompts can lead to outputs that feel less organic or literal in their interpretation.
Examples of Do's
- Forest ambience
- Chatter of people in a restaurant
- Room tone
- Traffic in a busy city
