US Patent 11223699 Multiple user recognition with voiceprints on online social networks

In one embodiment, a method includes, by one or more computing devices of an online social network, receiving, from a client system of a first user of the online social network, an audio input from a second user, wherein the audio input comprises one or more voice commands, identifying the second user based on a comparison of the audio input to one or more voiceprints stored by the online social network, wherein each voiceprint comprises audio data for auditory identification of a unique user of the online social network, determining a relationship status between the first user and the identified second user within the online social network, and determining whether to perform an action associated with each voice command based on permission settings associated with the action and the determined relationship status between the first user and the identified second user.