Apple is working on a feature to read text messages with the voice of their senders

Apple has patented a new technology that may soon reach iPhone users that allows them to listen to text messages received within the iMessage application with the voice of the message sender himself.

iPhone users can now ask Siri to read incoming text messages, as Siri’s smart assistant relies on Speech-to-text technology to perform this task. But instead of reading with Siri’s voice, the new technology promises to read texts using the sender’s own voice.

In order for this to be possible, the new feature requires both parties to the conversation to share an audio recording of each that includes some short phrases on which the new service will rely on using these samples to generate the required sound.

The patent explained how the technology works by saying: “When a text message arrives from a person named “John”, the user may wish to listen to the message in the voice of “John” himself, which improves the use experience and increases the efficiency of using the device by eliminating the need to read the received message. .

According to the text of the patent, the user records an audio sample that is stored in his device, and any other user can send a request to obtain the sample, and with the consent of the first user, the sample is sent to the phone of the second user, and once this process is completed, the first user is asked if he also wants to listen to messages In the voice of the other party, and in the event of approval, another exchange of the sample takes place from the second party, so that listening to messages in the voice of their owners becomes possible for both parties.

Through the settings, the sender can activate sending the audio sample automatically to any party to whom he sends a text message, or to specific previously selected destinations, in a way that the user does not have to perform this step manually every time.

It is noteworthy that the technology of converting text into speech using real human voices has been available for a few years and it relies on advanced artificial intelligence algorithms and deep learning techniques to create artificial voices that simulate human voices.

It is worth noting that patent registration does not necessarily mean the commercial launch of the registered features, and therefore it is not possible to predict whether the feature will see the light soon.