Respository with some of the models trained in ACMus publications.
For ensemle size detection best CNN model was taken from: Ensemble size classification in Colombian Andean string music recordings
For speech music classification best CNN model was taken from - refer to baseline on T3: Analyzing the potential of pre-trained embeddings for audio classification tasks
Models and commandline scripts are included for ensemble size classification and speech/music detection as well as example files. Further data can be found at Zenodo ACMUS-MIR
Install reuqired packages with pip or conda using the provided requirement.txt file for python 3.6.
Option A: Create new environment and install dependencies using conda:
# Create new env
$ conda create -n acmus_models python=3.6
# Activate it
$ conda activate acmus_models
# Install from requirements file
(acmus_models)$ pip install --user --requirement requirements.txt
Option B: Install using pip only:
$ pip install -r requirements.txt
Option C: Install main packages manually:
$ pip install tensorflow==1.15.2
$ pip install librosa==0.7.2
Either "ensemble_size_inference_main.py" or "speech_music_inference_main.py" for each task. Arguments are "-i" for input file or folder with files. These folders should contain only audio files! The prediction results are written to a csv file. The output file name can be set using "-o".
$ python ensemble_size_inference_main.py -i example_files/ -o output.csv
MIT License
Copyright (c) 2020 ACMus - Advancing Computational Musicology
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.