Disclosed embodiments provide techniques for suggesting a visual effect based on detected sounds. The sounds can be speech and/or music. Tempo and song identification techniques may be used to determine criteria for selecting visual effects to present to a user. The user selects a visual effect from the suggested visual effects and applies the visual effect to an image acquired by a camera. A modified image that combines the original acquired image with the visual effect is transmitted to another user during communication such as video chat, or, alternatively, the modified image may be posted to a social media account.