Beijing Xiaomi Pinecone Electronics Co., Ltd.

Guodu Cai

Yifeng Wang

Xin Wang

Peng Gao

Shuo Yang

Lihan Li

A training method includes: acquiring voice data of a user on a plurality of devices, and extracting acoustic features based on the voice data to construct a training data set; inputting the training data set into a neural network model, training the neural network model, and acquiring a loss function of each training of the neural network model; and determining that the training of the neural network model is completed when the loss function converges, wherein the loss function is a weighted sum of a classification loss function, a reconstruction loss function, a difference loss function, and a similarity loss function.

Method for training a voiceprint extraction model and method for voiceprint recognition, and device and medium thereof

A method for detecting an audio signal, the method comprises: obtaining a speech segment and a non-speech segment of an audio signal to be detected, extracting a first audio feature of the speech segment and a second audio feature of the non-speech segment, detecting the first audio feature using a predetermined speech segment detection model to obtain a first detection score, detecting the second audio feature using a predetermined non-speech segment detection model to obtain a second detection score, and determining whether the audio signal belongs to a target audio based on the first detection score and the second detection score.

Growing community of inventors

Beijing, China

Guodu Cai