Can analayze a wave file (or directory of wave files) of any sampling frquency and can save the following: - a csv file in the csv/ folder: Has three columns for time, amplitude, and speechflag. speechflag will be 1 if speech is detected in that sample - a text file (in /txt folder): has details of speech length and processing time. - a waveform in png format (in folder /png) (optional) has speech activity highlighted in red. This is optional and PLOT_SAVE variable needs to be set True inside audio_analysis.py - a binary file ( in /numpy folder) in numpy format (optional) Just run it with the command: python audio_analysis.py -i "inputfile.wav" to analyze a single file or python audio_analysis.py -d "inputdir" to analyze all wav files inside inputdir use option -s for a silent output where nothing will be printed on stdout