Audio voice and text synchronization method, editing method and editing system

Abstract

The invention discloses a voice and text synchronization method of an audio, an editing method and an editing system. The synchronization method comprises the steps of 1, decoding an audio file to acquire an audio waveform, and synchronously translating the audio into text; 2, performing interval division on the audio waveform; 3, computing an average relative level value of each interval; 4, acquiring a node recognition parameter of each interval according to the average relative level value of each interval; 5, acquiring the start moment, end moment and bottom noise area or mute area of eachsyllable in the audio according to the node recognition parameter of each interval and a preset node recognition standard, and sequentially numbering the same; and 6, building a moment synchronization table for syllables in text and voice, thus acquiring synchronization timestamps of the text and voice. According to the method provided by the invention, the audio editing efficiency can be greatlyimproved.

Description

technical field [0001] The invention belongs to the technical field of editing and editing voice and audio, and in particular relates to a method for synchronizing audio voice and text, an editing method and an editing system. Background technique [0002] Speech audio content editing of traditional radio stations and TV stations is mainly realized by using general-purpose audio editing software (such as audition software, samplitude software). Such software itself does not have a synchronous text editing window, and it needs to rely entirely on the auditory control of the human ear. Manual cutting and modification, the efficiency and accuracy are not high. Among them, the current audio manuscript is generally used to generate a document manuscript, and then the wps software and audition software are used to work continuously, and the audio is edited while modifying the file. The modification of the audio relies on manual monitoring, through the memory of the human brain, wo...

AI Technical Summary

Problems solved by technology

[0002] Speech audio content editing of traditional radio stations and TV stations is mainly realized by using general-purpose audio editing software (such as audition software, samplitude software). Such software itself does not have a synchronous text editing window, and it needs to rely entirely on the auditory control of the human ear. Manual cutting and modification, the efficiency and accuracy are not high

Among them, the current audio manuscript is generally used to generate a document manuscript, and then the wps software and audition software are used to work continuously, and the audio is edited while modifying the file. The modification of the audio relies on manual monitoring, through the memory of the human brain, word by word dictation and proofreading, that is, text deletion and adjustment and audio deletion and adjustment need to be completed in two steps. As a result, when major programs are broadcast, audio manuscripts and text manuscripts must be reviewed one by one due to the needs of review and release through different channels. When corresponding, the editing work will consume a lot of working time, and the working process is cumbersome and boring

Images