paulnasca / paulstretch_python Goto Github PK

Hi Paul,

I'm trying to run this on MacOS High Sierra 10.13.6. Python 2.7.10.

I've tried a variety of wav files from sample packs and so forth as a test, and have gotten nothing but "Error loading wav..."; I added a bit of code to get more information. Seems I get a ValueError exception, which contains the following: "string size must be a multiple of element size"

It is triggered from this line:

wavedata=scipy.io.wavfile.read(filename)

I can dig further but was hoping you might just have an idea of what might be wrong here.

All I did was clone, copy a wav to the working directory as a.wav, then run:

./paulstretch_stereo.py a.wav b.wav

Full output:

Paul's Extreme Sound Stretch (Paulstretch) - Python version 20141220
by Nasca Octavian PAUL, Targu Mures, Romania

stretch amount = 8
window size = 0.25 seconds
Error loading wav: a.wav
Traceback (most recent call last):
  File "./paulstretch_stereo.py", line 149, in <module>
    (samplerate,smp)=load_wav(args[0])
TypeError: 'NoneType' object is not iterable

Hi !

I am confused about which script to use ... because one says "newmethod", but "paulstretch_stereo" has been updated more recently !?
Anyways, thanks for those scripts, they are great :)

Hey!

I have no clue how to use this script. Maybe an example command would be nice to have.

Cheers!

how can i mimic the freeze effect?

Using python3, I cannot seem to use any of the argparse options or input a wav file. Can you give an example command-line syntax in the readme? I receive the same error trying to access the -help arg as well.

python3 paulstretch_newmethod.py TTLS.wav
print "Error loading wav: "+filename

Why randomize the phase? It makes everything sound super reverberant. Although I do enjoy it, and I must say it has become the signature paulstretch sound.

However, you get a cleaner extreme time stretch if you interpolate the change in phase from frame to frame. Interpolating phase itself won't produce meaningful results, and I know I was led astray initially by trying this, but interpolating the change in phase may sound more natural.

Check it out:
https://www.youtube.com/watch?v=h5SObE8JSX8
Read all about the technique here:
http://www.mitpressjournals.org/doi/pdf/10.1162/comj.2008.32.3.87