| Instruction | Generation Result |
|---|---|
| A telephone dialing tone followed by a plastic switch flipping | |
| A train horn blows and fades, then metal clacking occurs | |
| An insect buzzing as plastic clacks and plastic slaps a hard surface | |
| Heavy rainfall with a brief muffled thunder from outside | |
| Sanding and scraping followed by a man speaking | |
| The wind is blowing, and a person is whistling a tune | |
| Ambulance siren repeatedly and then continuous | |
| A toilet flushes and water drains |
| Task | Instruction | Reference Audio | Editing Result |
|---|---|---|---|
| Adding | Insert a sound of pouring water at the end | ||
| Put a sound of toilet flush at the midpoint | |||
| Mix a sound of bird chipping throughout the audio | |||
| Add a sneezing at the beginning of the audio | |||
| Removing | Drop laughing from the audio | ||
| Remove the sound of cat meowing | |||
| Replacement | Exchange clock tick with violin | ||
| Replace the sound of dog with sheep | |||
| Reordering | Reverse the order of pouring water and air conditioner | ||
| Swap the order of rain and munching | |||
| Inpainting | Inpaint the audio | ||
| Complete the missing segment | |||
| Super-resolution | Synthesize the missing high frequencies | ||
| Increase audio resolution |