chords-recognition

KO | EN

gitlite — search

#python #hacktoberfest #react #go #machine-learning #rust #linux #macos #java #cli #awesome-list #security #openai #open-source

chords-recognition

(★ 29)

No description available.

chords-recognition Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# General information
Current repository contains complete Chords Recognition system. System takes audiofile as input and returns timings for each chord according to [Mirex categories](https://www.music-ir.org/mirex/wiki/2019:Audio_Chord_Estimation) (See example bellow). System implementation is based on Recurrent neural net with LSTM cell. 
![output example](https://sun9-12.userapi.com/c856016/v856016534/de37d/YuIR9iuvQUU.jpg)
 
In addition, repo contains realtime demo and simple web-app, which allows user to apload song and get chords for it.
Demo video of realtime chords recognition is avaliable [here](https://www.youtube.com/watch?v=rxZA6Nwei4E) 
Project implemented as a part of Master thesis, full text avaliable [here](https://docs.google.com/document/d/1vFkN35JcqBUbSRvDcb89lD4Pxb6BW4Qk_xi3q33EgsU/edit?usp=sharing).
# Datasets
- [TheBeatles180 Isophonics dataset](http://www.isophonics.net/content/reference-annotations-beatles) 
 - [audio](https://drive.google.com/open?id=1WzdcHeLeFHrYu_2_NDTEEglfrKLLKc1c) - extract in data/audio
 - [gt](https://drive.google.com/open?id=1EK59lTqt6iXN7ykdZXJyrYLuHmws36lh) - extract in data
 - [converted](https://drive.google.com/open?id=1Yh4dBJqtYkN7Hy5qp8E-dKeITX9D38I_) - extract in data/converted
 - [converted with librosa](https://drive.google.com/open?id=1hiTi_CPKxu9Qpli-zch1vINa4iY5iE9s)
 - [converted_list](https://drive.google.com/open?id=1E-TVqZvlFIJ2KzxmkkdhPxlKXxzQAcZJ) - extract to root folder
- [CaroleKingQueen Isophonic dataset](http://www.isophonics.net/content/reference-annotations-carole-king)
 - [audio](https://drive.google.com/open?id=1GVBNRwZ_YFHD9aroqP_NaI57H07_R3GR)
 - [converted with librosa](https://drive.google.com/open?id=1WbyWA4UcYuMHw7QvXMpb-i1PJrpKDKiZ)
 - [converted list](https://drive.google.com/open?id=1vMFUvgJrAzCsO4PjZ98pYkirGRGNt3Sk)
- [USPop2002](https://labrosa.ee.columbia.edu/projects/musicsim/uspop2002.html)
 - [audio](https://drive.google.com/open?id=161eEk-o1ulujRh_n-hYmQxlwdmbhWWja)
 - [converted with librosa](https://drive.google.com/open?id=1w8Mo2r6ml1v76SiU3MTjgoWiNjFLOH9a)
 - [converted list](https://drive.google.com/open?id=1V5zdvcB50YLfnTlsKyZspvTuVbWghrLE)
- JayChou29
 - [audio](https://drive.google.com/open?id=1s55LgFKyybeSueruV8Xvtwh6TE4yJnAb)
- Full datasets
 - [Full list](https://drive.google.com/open?id=1m8wC0vAc4p-HbNx68PH1gOfSKv2FE_EU)
 - [Full list with modulation](https://drive.google.com/open?id=1HSWo6Wv1fWmWjViN13TERxIgUpCMxoHe)
 - [Full dataset mod -1, -3](https://drive.google.com/file/d/19kqa5sZ7YwWd4KHZ8DdyUSP1eDkKnhiL)
 - [Full dataset mod 2, 4](https://drive.google.com/file/d/19kqa5sZ7YwWd4KHZ8DdyUSP1eDkKnhiL)
# Structure of repo:
- data
 - audio - default folder for raw audio files
 - converted - default folder for the saving preprocessed data in csv format
 - gt - contains .lab files with chords
 - tracklists - contains lists of paths to audio files starting with 'audio_root' parameter
 - predicted - contains predictions for TheBeatles180 and JayChou29 datasets, reports generated by MusOOEvaluator and pretrained models
# Installation
Download raw and converted datasets
```
bash ./data/download.sh
```
# Preprocessing
System computes notegramms (252 bins per sample) as described in Mauch 2010 (p.98) with hop_length=512, sample_rate=11025, window_size=4096
### How to use:
```
python preprocess.py --songs_list data/tracklists/TheBeatles180List
```
#### Parameters:
--songs_list 
--audio_root, default: data/audio/ 
--gt_root, default: data/gt/ 
--conv_root, default: data/converted/, determines folder where converted datasets will be saved 
--subsong_len, default: 40, length of song part in seconds to be splitted during preprocess 
--song_len, default: 180 if subsong_len is not specified, song will be cutted or zeropaded to song_len 
--use_librosa, default: True, by default librosa.cqt will be used for preprocessing 
--songs_list, required, examples could be found in data/tracklists 
--num_bins, default: 84, defines number of bins during cqt 
--modulation_steps, list, default: [0], defines amount of modulation steps during CQT-preprocessing 
# Models:
## LSTM
### How to use:
```
python train_rnn.py --model LSTM --conv_list TheBeatles180List_converted_librosa.txt
```
#### Optional parameters:
--num_epochs, default: 2 
--learning_rate, default: 0.01 
--weight_decay, default: 1e-5 
--songs_list, default: data/tracklists/TheBeatles180List 
--audio_root, default: data/audio/ 
--gt_root, default: data/gt/ 
--conv_root, default: data/converted/, determins folder where converted datasets will be saved 
--conv_list, if specified, converted audio from list will be used for fitting model and convertation process will be skipped 
--category, default: MirexRoot 
--subsong_len, default: 40, length of song part in seconds to be splitted during preprocess 
--song_len, default: 180 if subsong_len is not specified, song will be cutted or zeropaded to song_len 
--hidden_dim, default: 200 
--num_layers, default: 2 
--batch_size, default: 4 
--sch_step_size, default:100, scheduler step size 
--sch_gamma, default:100, scheduler's gamma 
--, default:10, test model on train and val datasets every n iterations 
--use_librosa, default: True 
--save_model_as, if specified, model will be saved in pretrained folder 
#### Test
```
python test_nn.py --model pretrained/LSTM_MirexRoot_TheBeatles180_librosa.pkl --conv_root data/converted/librosa --conv_list TheBeatles180List_converted_librosa.txt
```
#### Results
##### Isophonic 2009
|Model| MirexRoot | MirexMajMin | MirexMajMinBass | MirexSevenths | MirexSeventhsBass |
| ------------- | ------------- | ------------- | ------------- | ------------- | ------------- |
|BiLSTM|95.58% | 94.59% | 94.27% | 90.86% | 90.6% |
|BiLSTM with modulation| 95.31% | 93.43% | 94.98% | 91.67% | 91.25% |
|BiGRU with modulation| 93.69% | 91.30% | 90.75% | 89.46% | 87.43% |

##### JayChou
|Model| MirexRoot     | MirexMajMin  | MirexMajMinBass | MirexSevenths | MirexSeventhsBass |
| ------------- | ------------- | ------------- | ------------- | ------------- | ------------- |
|BiLSTM|67.7%  | 61.42%  | 59%  | 41%  | 39.33%  |
|BiLSTM with modulation| 72.04%  | 69.58%  | 66.09%  | 48.97%  | 44.98%  |
|BiGRU with modulation| 68.59%  | 65.27%  | 62.59%  | 45.08%  | 40.37%  |

## Random forest
 Accuracy on test-set: Mirex_Root:55% 
### How to use:
#### Train
```
python train_rf.py
```
##### Optional parameters:
--songs_list, default: data/tracklists/TheBeatles180List 
--audio_root, default: data/audio/ 
--gt_root, default: data/gt/ 
--conv_root, default: data/converted/, determins folder where converted datasets will be saved 
--conv_list, if specified, converted audio from list will be used for fitting model and convertation process will be skipped 
--category, default: MirexRoot 
--subsong_len, default: 40, length of song part in seconds to be splitted during preprocess 
--song_len, default: 180 if subsong_len is not specified, song will be cutted or zeropaded to song_len 
--criterion, default: entropy 
--max_features, default: log2 
--n_estimators, default: 1 
#### Test
```
python test_rf.py --model pretrained/RF_MirexRoot_TheBeatles180_librosa.pkl --conv_root data/converted/librosa --conv_list TheBeatles180List_converted_librosa.txt
```
##### Optional parameters:
--songs_list, default: data/tracklists/TheBeatles180List 
--audio_root, default: data/audio/ 
--gt_root, default: data/gt/ 
--conv_root, default: data/converted/, determins folder where converted datasets will be saved 
--conv_list, if specified, converted audio from list will be used for fitting model and convertation process will be skipped 
--category, default: MirexRoot 
--subsong_len, default: 40, length of song part in seconds to be splitted during preprocess 
--song_len, default: 180 if subsong_len is not specified, song will be cutted or zeropaded to song_len

# Applications
## Realtime demo 
```
python realtime.py --category MirexRoot
```
## Web-application for chords recognition
```
python web_app\server.py
```

chords-recognition

chords-recognition Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

chords-recognition

chords-recognition Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README