Audio processing method, audio processing device and readable storage medium

Abstract

The invention provides an audio processing method, an audio processing device and a readable storage medium. The audio processing method comprises the following steps: separating a plurality of effective voice segments from an obtained original audio; performing energy detection on each effective voice segment, and determining at least one first qualified voice segment of which the energy is within a standard energy range; based on the signal-to-noise ratio corresponding to each first qualified voice segment, determining at least one second qualified voice segment of which the signal-to-noiseratio is higher than a standard signal-to-noise ratio; and based on the speech speed of each second qualified speech segment, screening out at least one target speech segment of which the speech speedis within a preset speech speed range from the at least one second qualified speech segment, so that a plurality of target voice segments can be screened from the original audio, and the target voicesegments are directly processed in the subsequent voice processing process, the voice processing quantity is reduced, and the voice processing efficiency is improved.

Description

technical field [0001] The present application relates to the technical field of speech processing, in particular to an audio processing method, an audio processing device and a readable storage medium. Background technique [0002] With the rapid development of speech recognition technology, related technologies for speech recognition have emerged as the times require. Speech recognition can be applied to the service industry to collect the voices of offline staff and analyze the voices of staff to determine the The work of the staff is carried out, so as to solve the problems in the service process in a targeted manner and improve customer satisfaction. [0003] At this stage, staff generally serve customers directly in offline stores. When collecting the voices of staff during the service process, the collected audio usually contains a large number of voice fragments, a large amount of noise, or a large number of silent voice fragments. When processing the collected audi...

AI Technical Summary

Problems solved by technology

[0003] At this stage, staff generally serve customers directly in offline stores. When collecting the voices of staff during the service process, the collected audio usually contains a large number of voice fragments, a large amount of noise, or a large number of silent voice fragments. When processing the collected audio, the amount of data processing is too large, resulting in low audio processing efficiency

Images