Below you will find examples for plotting:
- Short audio clips (~1s)
- Time domain plot short.r
- Audio 0.8s.wav
- Long audio clips (10s+)
- Time domain plot long.r
- Try your own WAV file!
- Vibration data (ms)
- Impulse plot.r
- Impulse data.txt
Example scripts and associated data files can be found here.
Short Audio Clips (~1s)
We will use the package tuneR to read in a wav file and convert it to a series of amplitude data.
Read in the wav file using a dialog selection box:
myFile <- file.choose() sndObj <- readWave(myFile,header=FALSE)
Check the properties of the wav file using:
str(sndObj)
This will show you information such as the number of channels, sample rate, and bit depth.
Now we need to build a data frame that contains one column of time information and one of amplitude. First we’ll isolate the left channel:
s1 <- sndObj@left
Now we’ll convert the values from bit steps to values between -1 and 1:
s1 <- s1 / 2^(sndObj@bit -1)
Next we need to create a column of time values:
filelength <- length(s1) timeArray <- (0:(filelength-1)) / [email protected] timeArray <- timeArray * 1000 #scale to milliseconds
Now we can create the data frame and carry on as in the previous example.
mydata <- data.frame(timeArray,s1)
Long Audio Clips (10s+)
Due to the large number of samples that are being processed when creating a time domain plot the export time increases dramatically with file length. For example, a data set containing 1,000,000 samples ([email protected]) takes 1min to export. To reduce this time (at the expense of accuracy) we can throw some samples away:
mydataextract <- mydata[c(TRUE,rep(FALSE,3)), ]
This will extract the rows of our choosing. Using 3 will extract rows 1, 5, 10, 15 etc. Be careful not remove too many rows as this will make a very noticeable difference in your plots.
Remember to remove line:
timeArray <- timeArray * 1000 #scale to milliseconds
As well as editing the yaxis formatting when plotting in seconds.
Vibration Data (ms)
The plot below shows an impulse created from an impact hammer and the subsequent response from an accelerometer. The example data is provided in csv format.
