Abstract: The present disclosure discloses an address information feature extraction method based on a deep neural network model. The present disclosure uses a deep neural network architecture, and transforms tasks, such as text feature extraction, address standardization construction and semantic-geospatial fusion, into quantifiable deep neural network model construction and training optimization problems. Taking a character in an address as a basic input unit, the address language model is designed to express it in vectors, then a key technology of standardization construction of Chinese addresses is realized through neural network target tasks.

Abstract: A call processing method and an electronic device are provided. The method is performed by an electronic device and includes: in a case that a call is connected successfully, obtaining the number of speakers in the current call. The method further includes: when the number of speakers is greater than a preset threshold, determining whether a user of the electronic device is a target user. The method additionally includes: if the user of the electronic device is a non-target user, disabling a speaking channel of the electronic device; and if the user of the electronic device is a target user, keeping the speaking channel of the electronic device on.

Abstract: The present disclosure discloses an address information feature extraction method based on a deep neural network model. The present disclosure uses a deep neural network architecture, and transforms tasks, such as text feature extraction, address standardization construction and semantic-geospatial fusion, into quantifiable deep neural network model construction and training optimization problems. Taking a character in an address as a basic input unit, the address language model is designed to express it in vectors, then a key technology of standardization construction of Chinese addresses is realized through neural network target tasks.