basic_audioqc_example

#!/bin/bash

#makes a folder called qccheck inside the folder you run the script on
mkdir qccheck
# makes a loop that will run on every .wav file in the folder. It takes every .wav it finds and uses that as the variable i
for i in *.wav ; do
    #makes a variable that will be used for the name of the qc metadata dump text files. Uses variable i (name of input file), for example test.wav would turn into test.wav_qc.txt
    qcreport="qccheck/${i}"_qc.txt
    #writes a heading for the csv to the metadata dump file
    echo 'frame,time_stamp,peak_db' > "${qcreport}"
    #this is the ffprobe command that outputs all the metadata to the text file
    ffprobe -f lavfi -i amovie="${i}",astats=metadata=1:reset=1 -show_entries frame=pkt_pts_time:frame_tags=lavfi.astats.Overall.Peak_Level -of csv >> "${qcreport}"
done

#moves into the qccheck folder to analyse the results of the metadata dump
cd qccheck

#makes a loop that runs on every .txt file in the qccheck folder
for i in *.txt ; do
    # This starts a 'read' loop that takes every line in the file and assigns it to a variable called audiodata
    while read audiodata ; do
        #this makes a variable that contains the peak db data from each line by using the 'cut' command set to deliminate on ',' (so it takes the third value from the CSV lines)
        peakdB=$(echo "${audiodata}" | cut -d ',' -f3)
        #this looks ugly because bash doesn't like decimals so we have to 'pipe' our number comparison ( | is a pipe) into a command called bc. Long story short, it checks to see if the peakdB variable has a value higher than -0.3, and if it does, it executes the following steps)
        if [ $(echo "${peakdB} >= -0.1" | bc) -ne 0 ] ; then
            echo "DANGER $i contains high audio levels"
            echo "${audiodata}"
        fi
    done < "${i}"
done