carlsouthall / adtlib Goto Github PK
View Code? Open in Web Editor NEWAutomated Drum Transcription Library
License: BSD 2-Clause "Simplified" License
adtlib's People
Contributors
Stargazers
Watchers
Forkers
python3pkg sudara lflee samikshan ruohoruotsi rockyplum bbqbrew keunwoochoi radkoff xty438307820 jarovaisanen tatsubori shirleyhywl audio-modeling-tools katpasalo benedekpasztor devnello fdch ismaildalgatov gary109 au-deps baukaalm sitongye p3zo leehyobeom brunoimbrizi geraher eboateng asantos-6 jjkimmadtlib's Issues
IndexError with some audio clips
Hi there,
Just for some wav files I'm getting the error:
INFO:tensorflow:Restoring parameters from /Users/ignaciot/miniconda3/lib/python3.6/site-packages/ADTLib/files/SnareADTLibAll
INFO:tensorflow:Restoring parameters from /Users/ignaciot/miniconda3/lib/python3.6/site-packages/ADTLib/files/HihatADTLibAll
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/Users/ignaciot/miniconda3/lib/python3.6/site-packages/ADTLib/__init__.py", line 24, in ADT
Peaks.append(utils.meanPPmm(AFs[j][:,0],PP[j,0],PP[j,1],PP[j,2]))
File "/Users/ignaciot/miniconda3/lib/python3.6/site-packages/ADTLib/utils/__init__.py", line 44, in meanPPmm
np.delete(onsets,onsets[i-1+ind])
File "<__array_function__ internals>", line 6, in delete
File "/Users/ignaciot/miniconda3/lib/python3.6/site-packages/numpy/lib/function_base.py", line 4406, in delete
keep[obj,] = False
IndexError: arrays used as indices must be of integer (or boolean) type
Any hints of what might be causing this? These are short clips from songs with a very clearly defined beat pattern.
Input arrays must have same number of dimensions
Thanks for releasing this library!
I was trying it out first with some wave files I had, which gave me no problems. I later tried to convert an mp3 to wav using ffmpeg (mono, 44.1KHz):
ffmpeg -i in.mp3 -acodec pcm_s16le -ac 1 -ar 44100 out.wav
When using default settings I get the following error:
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-41-50059aaecf0f> in <module>()
----> 1 out=ADTBDRNN(['/home/cuaras/out.wav'])
/home/cuaras/Documents/workspace/drum/venv/lib/python2.7/site-packages/ADTLib/models/__init__.pyc in ADTBDRNN(TrackNames, out_sort, ret, out_text, savedir, close_error, lambd)
83 P[j][i]=peak
84
---> 85 P=arrange_output(P,output_sort=out_sort)
86
87 if out_text == 'yes':
/home/cuaras/Documents/workspace/drum/venv/lib/python2.7/site-packages/ADTLib/utils/__init__.pyc in arrange_output(Inputs, output_sort)
44 Out[i][j][k][1]=Names[j]
45
---> 46 Out[i]=np.concatenate([Out[i][0],Out[i][1],Out[i][2]],axis=0)
47
48 if output_sort=='time':
ValueError: all the input arrays must have same number of dimensions
Maybe it's some configuration error on my part. Here is the link to the file:
https://dl.dropboxusercontent.com/u/12497232/out.wav
ValueError: setting an array element with a sequence.
An input of one filename as a singleton list works, but a list of two filenames raises an error.
Good input:
ADTBDRNN(['simple_loop.wav'])
Output:
[array([ 0. , 0.04643991, 0.52244898, 0.52244898, 0.98684807,
0.99845805, 1.04489796, 1.24226757, 1.24226757, 1.30031746,
1.52090703, 1.52090703, 1.59056689]),
array(['BD', 'HH', 'SD', 'HH', 'BD', 'HH', 'BD', 'BD', 'HH', 'BD', 'SD',
'HH', 'HH'],
dtype='|S32')]
Good input:
ADTBDRNN(['125_bounce.wav'])
Output:
[array([ 0.49922902, 0.510839 , 0.58049887, 0.95201814, 0.95201814,
1.45124717, 1.53251701, 2.31038549, 2.41487528, 2.47292517,
2.80961451, 3.36689342, 3.44816327, 3.81968254, 3.81968254,
4.23764172, 4.34213152, 4.42340136, 5.09678005, 5.29414966,
5.36380952, 6.33904762, 6.6060771 , 7.0124263 , 7.05886621,
7.20979592, 7.27945578, 7.43038549]),
array(['SD', 'HH', 'HH', 'HH', 'BD', 'SD', 'HH', 'BD', 'SD', 'HH', 'BD',
'SD', 'HH', 'HH', 'SD', 'BD', 'SD', 'HH', 'BD', 'SD', 'HH', 'HH',
'BD', 'BD', 'SD', 'SD', 'SD', 'HH'],
dtype='|S32')]
Bad Input:
ADTBDRNN(['simple_loop.wav', '125_bounce.wav'])
Output:
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-8-88f3a1bbf5de> in <module>()
----> 1 ADTBDRNN(['simple_loop.wav', '125_bounce.wav'])
/Library/Python/2.7/site-packages/ADTLib/models/__init__.pyc in ADTBDRNN(TrackNames, out_sort, ret, out_text, savedir, close_error, lambd)
85 P[j][i]=peak
86
---> 87 P=arrange_output(P,output_sort=out_sort)
88
89 if out_text == 'yes':
/Library/Python/2.7/site-packages/ADTLib/utils/__init__.pyc in arrange_output(Inputs, output_sort)
40
41 Out[i]=list(np.zeros(len(Inputs[i])))
---> 42 Out1[i]=list(np.zeros((1,2)))
43 for j in xrange(len(Inputs[i])):
44 Out[i][j]=list(np.zeros((len(Inputs[i][j]))))
ValueError: setting an array element with a sequence.
Website upload doesn't work ?
Hi!
I wanted to test your library but seems that the demo website you made can't upload file and return error 500 👍 https://dmtlab.bcu.ac.uk/ADT/
Also, the usage page you provided in your readme is not available ?
https://github.com/CarlSouthall/ADTLibNew/blob/master/usage.md
Thanks!
Hi, I have some question regarding the datasets
I have started automatic drum transciption, but I'm lacking the datasets.
I'd like to ask where I can obtain the datasets for training.
Thanks a lot.
Only half of the song is processed?
Hi Carl,
I was trying to run your algo on couple of songs but I've noticed whatever I pass as input I will always get only half of the song processed. For instance if I pass 20 seconds in input I can see only BD, SD and HH up to 10 seconds same if I double those values. Any idea of why this might be happening?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.