Abstract: A method and server for generating content for a user are disclosed. The method includes generating, via a first model, a first output for the user. The first output is representative of a predicted item of a first type. The method includes ranking, via a second model, the items of the second type into a ranked list based on the first output. A rank of an item is indicative of a likelihood of that the user interacts with the given item of the second type if the user performs an event of the pre-determined type on the given predicted item of the first type. Content is then selected from the ranked list and transmitted to a user device.

Abstract: Methods and servers for displaying a digital item to users of a recommendation system are disclosed. The method comprises determining first and second recommendable contents that include the digital item for a first and a second user of a first and a second electronic devices respectively. The digital item is associated with a plurality of actions. The method comprises triggering display of at least some content from the first and second recommendable content including the digital item on the first and second electronic devices respectively. The digital item is associated with a first action triggerable by the first user on the first electronic device and a second action triggerable by the second user on the second electronic device. The first and second actions are selected among the plurality of actions based information about the first and second users respectively.

Abstract: Method and servers for generating a speech model for generating signals representative of utterances in a first language based on signals representative of utterances in a second language are disclosed. The method comprises transmitting a first and a second speech models to a first and a second devices of a first and a second users respectively. The first device is communicatively coupled with the second device by an encrypted communication link. A third speech model is acquired from the second device based on a local training of the second speech model on the second device. A training set comprises a first and a second decrypted signals representative of an utterance of the first user in the first language and a translated utterance of the first user in the second language respectively. The speech model is locally generated by the server by combining the second and third speech models.

Abstract: A method and server for generating content for a user are disclosed. The method includes generating, via a first model, a first output for the user. The first output is representative of a predicted item of a first type. The method includes ranking, via a second model, the items of the second type into a ranked list based on the first output. A rank of an item is indicative of a likelihood of that the user interacts with the given item of the second type if the user performs an event of the pre-determined type on the given predicted item of the first type. Content is then selected from the ranked list and transmitted to a user device.

Abstract: Methods and servers for displaying a digital item to users of a recommendation system are disclosed. The method comprises determining first and second recommendable contents that include the digital item for a first and a second user of a first and a second electronic devices respectively. The digital item is associated with a plurality of actions. The method comprises triggering display of at least some content from the first and second recommendable content including the digital item on the first and second electronic devices respectively. The digital item is associated with a first action triggerable by the first user on the first electronic device and a second action triggerable by the second user on the second electronic device. The first and second actions are selected among the plurality of actions based information about the first and second users respectively.

Abstract: Method and servers for generating a speech model for generating signals representative of utterances in a first language based on signals representative of utterances in a second language are disclosed. The method comprises transmitting a first and a second speech models to a first and a second devices of a first and a second users respectively. The first device is communicatively coupled with the second device by an encrypted communication link. A third speech model is acquired from the second device based on a local training of the second speech model on the second device. A training set comprises a first and a second decrypted signals representative of an utterance of the first user in the first language and a translated utterance of the first user in the second language respectively. The speech model is locally generated by the server by combining the second and third speech models.