Data processing method and device, electronic equipment and readable medium

Abstract

The embodiment of the invention provides a data processing method and device, electronic equipment and a readable medium. The method comprises the following steps: inputting feature data corresponding to sequence data into a neural network; wherein a plurality of network layers of the neural network are divided into a plurality of network layer groups according to a processing sequence, according to the total frame number of sequence data corresponding to input feature data, processing of the network layer groups corresponding to the total frame number is started one by one according to the processing sequence, and difference values between the total frame numbers corresponding to every two adjacent network layer groups are the same or different. According to the method, the calculation of the neural network is split, the calculation is shared in a longer time, the problem of peak of calculation resource occupation when a plurality of frames are input each time is avoided, the problems of overhigh calculation resource load at a local moment and overlow calculation resource load at most moments are solved, the load peak value is reduced, the power consumption of the whole calculation is reduced, and the calculation efficiency is improved. The response speed of calculation is accelerated, and meanwhile waste of calculation resources is reduced.

Description

technical field [0001] The present application relates to the technical field of data processing, and in particular to a data processing method, a data processing device, electronic equipment, and a machine-readable medium. Background technique [0002] The sound of audio playback is generated over time, so it is naturally represented as a one-dimensional time series, and such data is called sequence data. Words in the language also appear one by one, such data is also sequence data, and they all contain the concept of time. The method of converting such a sequence of data into a fixed-length vector representation is called sequence modeling (Sequence Models). [0003] At present, in the sequence modeling of audio data, the usual calculation process is: the audio signal is framed and windowed and features are extracted, and features of several frames (for example, W=100 frames) are formed into a sliding window, and as a whole, input a The convolutional neural network perfo...

AI Technical Summary

Problems solved by technology

[0005] In view of the above problems, the embodiments of the present application propose a data processing method, data processing device, electronic equipment, and machine-readable medium that overcome the above problems or at least partially solve the above problems. The embodiments of the present application can solve the problem of unbalanced computing resource occupation The problem

Images