Merge pull request #51 from SuperKogito/edit-paper

Edit comparison to existing libraries

README.md

@@ -115,6 +115,12 @@ cd spafe
 python setup.py install
 ```
 
+## Why use Spafe? 
+
+Unlike most existing audio feature extraction libraries ([python_speech_features](https://github.com/jameslyons/python_speech_features), [SpeechPy](https://github.com/astorfi/speechpy), [surfboard](https://github.com/novoic/surfboard) and [Bob](https://gitlab.idiap.ch/bob)), Spafe provides more options for spectral features, notably, Bark Frequency Cepstral Coefficients (BFCCs), Constant Q-transform Cepstral Coefficients (CQCCs), Gammatone Frequency Cepstral Coefficients (GFCCs), Power-Normalized Cepstral Coefficients (PNCCs), Phase based Spectral Root Cepstral Coefficients (PSRCCs) extraction algorithms. 
+Most existing libraries and to their credits provide great implementations for features extraction but are unfortunately limited to the Mel Frequency Features (e.g. MFCC) and at best have Bark frequency and linear predictive coefficients additionally. [Librosa](https://github.com/librosa/librosa) for example includes great implementation of various algorithms (only MFCC and LPC are included), based on the Short Time Fourrier Transform (STFT), which is theoretically more accurate but slower than the Discret Fourrier Transform used in Spafe's implementation.
+
+
 ## How to use
 
 Various examples on how to use spafe are present in the documentation [https://superkogito.github.io/spafe](https://superkogito.github.io/spafe).

paper/paper.bib

@@ -69,6 +69,16 @@ @article{speechpy:2018
    journal = {Journal of Open Source Software}
  }
 
+@InProceedings{librosa:2015,
+  author    = {{B}rian {M}c{F}ee and {C}olin {R}affel and {D}awen {L}iang and {D}aniel {P}.{W}. {E}llis and {M}att {M}c{V}icar and {E}ric {B}attenberg and {O}riol {N}ieto},
+  title     = {librosa: {A}udio and {M}usic {S}ignal {A}nalysis in {P}ython},
+  booktitle = {{P}roceedings of the 14th {P}ython in {S}cience {C}onference},
+  pages     = {18-24},
+  year      = {2015},
+  editor    = {{K}athryn {H}uff and {J}ames {B}ergstra},
+  doi       = {10.25080/Majora-7b98e3ed-003}
+}
+
 @book{python_speech_features:2020,
   title={jameslyons/python_speech_features: release v0.6.1},
   url={https://zenodo.org/record/3607820},

paper/paper.md

@@ -26,7 +26,7 @@ In speech processing, features extraction is essentially the estimation of a par
 This is a key step in any audio based modeling and recognition process (e.g. speech recognition, sound classification, speaker authentication etc.).
 There are several speech features to extract, such as the Linear Frequency Cepstral Coefficients (LFCC), Mel Frequency Cepstral Coefficients (MFCC), Linear Predictive Coding (LPC), and Constant-Q Cepstral Coefficients (CQCC) etc.
 Each type of features has its own advantages and drawbacks (e.g. noise robustness, complexity, inter-components correlation etc.) that can directly affect the researched topic.
-Unfortunately, existing libraries for extracting these features (e.g. python_speech_features [@python_speech_features:2020], SpeechPy [@speechpy:2018] and Bob [@bob:2017]) are limited and mostly focus on one extraction technique (e.g. MFCC), thus it is hard to find reliable implementations of other features extraction algorithms.
+Unfortunately, existing libraries for extracting these features (e.g. librosa [@librosa:2015], python_speech_features [@python_speech_features:2020], SpeechPy [@speechpy:2018] and Bob [@bob:2017]) are limited and mostly focus on one extraction technique (e.g. MFCC), thus it is hard to find reliable implementations of other features extraction algorithms.
 Consequently, this slows down the research and hinders the possibility of exploring and comparing these different approaches.
 Hence, the need for **spafe**, a straightforward solution that unites all these different techniques in one python package.