bp/documents
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Initial document commit. Added CV, Hud research proposal and QMUL statement to repo

Browse Source
This commit is contained in:
Pezz89
2016-03-29 13:46:54 +01:00
commit 4149f418ae
3 changed files with 489 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CV/Sam_perry-CV.tex
+217
View File
@@ -0,0 +1,217 @@
\documentclass[10pt,letterpaper]{article}
\usepackage{hyperref}
\usepackage{geometry}
\usepackage{enumitem}
\usepackage{multicol}
\newcommand{\tabitem}{~~\llap{\textbullet}~~}
% Fonts
\usepackage[T1]{fontenc}
\usepackage[urw-garamond]{mathdesign}
% Set your name here
\def\name{Samuel Perry}
% The following metadata will show up in the PDF properties
\hypersetup{
colorlinks = true,
urlcolor = black,
pdfauthor = {\name},
pdfkeywords = {DSP, Programmer},
pdftitle = {\name: Curriculum Vitae},
pdfsubject = {Curriculum Vitae},
pdfpagemode = UseNone
}
\geometry{
body={6.5in, 9.0in},
left=1.0in,
top=1.0in
}
% Customize page headers
\pagestyle{myheadings}
\markright{\name}
\thispagestyle{empty}
% Custom section fonts
\usepackage{sectsty}
\sectionfont{\rmfamily\mdseries\Large}
\subsectionfont{\rmfamily\mdseries\itshape\large}
% Other possible font commands include:
% \ttfamily for teletype,
% \sffamily for sans serif,
% \bfseries for bold,
% \scshape for small caps,
% \normalsize, \large, \Large, \LARGE sizes.
% Don't indent paragraphs.
\setlength\parindent{0em}
% Make lists without bullets and compact spacing
\renewenvironment{itemize}{
\begin{list}{}{
\setlength{\leftmargin}{1.5em}
\setlength{\itemsep}{0.25em}
\setlength{\parskip}{0pt}
\setlength{\parsep}{0.25em}
}
}{
\end{list}
}
\setlist[enumerate]{itemsep=0.25em}
\begin{document}
% Place name at left
{\huge \name}
% Alternatively, print name centered and bold:
%\centerline{\huge \bf \name}
\bigskip
\begin{minipage}[t]{0.495\textwidth}
20 Lower Luton Road\\
Wheathampstead\\
Hertfordshire\\
AL4 8QZ\\
\end{minipage}
\begin{minipage}[t]{0.495\textwidth}
Phone: (+44) 7766 521596\\
Email: \href{mailto:samuel.perry89@gmail.com}{samuel.perry89@gmail.com} \\
Linked-in: \\\href{https://uk.linkedin.com/in/sam-perry-04245438}{https://uk.linkedin.com/in/sam-perry-04245438}
\end{minipage}
\section*{Personal Profile}
A highly motivated and ambitious graduate with a background in programming and
digital signal processing in a musical context. Aiming to further knowledge and
understanding of digital signal processing techniques, building on previous
experience in this area. Capable of understanding and utilising signal
processing techniques for the realization of signal processing applications for
a variety of use cases, as proved through recent studies and employment.
Through further studies in a technically oriented environment, the objective
is to gain a deeper understanding in this field in order to facilitate future
employment or research opportunities.
\section*{Employment}
\begin{itemize}
\item Institut de Recherche et Coordination Acoustique/Musique (IRCAM)
\end{itemize}
\subsection*{IRCAM}
Role: Student Research Assistant \\
Team: Analysis \& Synthesis team \\
Location: Paris, France \\
Period: August 2014 - July 2015 \\
\newline
Description: \\
Worked on a range of DSP related projects and tasks for the Analysis and
Synthesis team. Modified and improved a number of programs, primarily in
Python, with particular focus on vocal and musical processing. Major
project involved using audio descriptor analyses to drive transformations
on vocal corpus. Worked alongside a variety of researchers and
professionals developing new and innovative signal processing techniques in
fields of research such as such as vocal transformations and audio/musical
content analysis. \\
\newline
Key areas explored:
\begin{itemize}
\item \textit{Audio content analysis}\\
Utilised a number of audio descriptors to test for similarities in
audio for a content matching algorithm
\item \textit{Vocal segmentation/classification}\\
Improved the efficiency of the content matching algorithm through
addition of vocal segment classification and tree search algorithm.
\item \textit{Distributed computing/Asynchronous processing}\\
Debugged and improved a program that utilised distributed task
scheduling for computation heavy analysis of audio
\end{itemize}
\newpage
\section*{Education}
\begin{itemize}
\item Music Technology (BA), The University of Huddersfield, 2012.
\begin{itemize}
\item \emph{Final Research Project:} ``Audio Descriptor Driven
Concatenative Synthesis of Corpus Databases'' \\
- details of which can be found at:
\href{http://pezz89.github.io/pysound/}{http://pezz89.github.io/pysound/}.
\item \emph{Predicted Classification:} "Borderline 2:1/first" - refer to
A.Harker written reference.
\end{itemize}
\item Music Technology BTEC Extended Diploma.
\begin{itemize}
\item \emph{Achievement:} Triple distinction awarded.
\end{itemize}
\end{itemize}
\subsection*{Music Technology (BA)}
Overview: \\
Study involved developing a broad understanding of musical signal
processing techniques through modules in topics such as DSP, Interactive
Sound Design and a final research project based on a novel technique for
audio synthesis \\
A detailed understanding of signal processing methods such as signal filtering,
spectral and temporal analysis, and granular synthesis were developed
through the practical application of these techniques for creative
purposes. Developer environments and languages such as Matlab, Python and
C++ were used to apply these concepts in software. An understanding of
application in hardware was also developed through the use of
microcontrollers to develop implementations of digital filters.
\section*{Key Skills}
\textit{Competent in the following programming languages, packages and environments:}
\begin{multicols}{3}
\begin{itemize}
\item Python
\item Matlab
\item C++
\item \LaTeX
\item Max/MSP
\item Vim
\item Bash script
\item Mac OSX
\item Git
\item HDF5 File system
\item Unix
\end{itemize}
\end{multicols}
\section*{References}
\begin{table}[h]
\centering
\label{my-label}
\begin{tabular}{ll}
\textit{Employer Reference} & \textit{Academic Reference} \\
\begin{tabular}[c]{@{}l@{}}
Axel Roebel \\
Head of the Analysis/Synthesis Research Team \\
IRCAM \\
Research Institute, Paris \\
Contact details available on request. \\
\end{tabular} & \begin{tabular}[c]{@{}l@{}}
Alex Harker \\
Huddersfield University Lecturer \\
The University of Huddersfield \\
University Telephone: 01484 473043 \\
E-mail: a.harker@hud.ac.uk \\
\end{tabular}
\end{tabular}
\end{table}
% Footer
\bigskip
{\small Last updated: \today}
\end{document}
Hud_masters_proposal/Sam_Perry-Research_masters_proposal.tex
+131
View File
@@ -0,0 +1,131 @@
\documentclass{scrartcl}
\usepackage{enumitem}
\usepackage[british]{babel}
\usepackage[style=apa, backend=biber]{biblatex}
\DeclareLanguageMapping{british}{british-apa}
\usepackage{url}
\usepackage{float}
\restylefloat{table}
\usepackage{perpage}
\MakePerPage{footnote}
\addbibresource{~/PerryPerrySource/LaTeX/Hud_masters.bib}
\begin{document}
\title{Huddersfield Research Masters}
\subtitle{Combined F0 Estimation Algorithm Proposal}
\author{Sam Perry}
\date{}
\maketitle
\begin{abstract}
The pitch of audio is a perceptually important characteristic as it
forms the building block for musical characteristics such as key,
melody, and harmony. Many methods have been developed for estimating
the fundamental frequency of a signal, however very few have come close
to generating estimations that resemble human perception of pitch with
the same level of detail. Factors such as noise and the absence of a
clear fundamental frequency cause erroneous results in algorithms and
the concept of polyphony further complicates the problem as this
requires the separation of different notes. The variety of estimation
algorithms available has lead to a selection of methods that each
perform to varying standards depending on conditions. For example,
time-domain approaches, such as the autocorrelation approach, are able
to detect the correct pitch of a signal more accurately than frequency
domain approaches, such as the harmonic-product spectrum method, when
the fundamental frequency is missing. However neither is able to detect
multiple pitches in the way that the MUSIC algorithm can.
\end{abstract}
\section{Overview}
This project would aim to explore the possibility of combining pre-existing
algorithms based on audio descriptor analyses in order to adaptively select
the algorithm with the best chance of an accurate estimate. The aim would
be to create a robust tool for offline (and potentially realtime)
estimation of F0 values.
\section{Background/Existing Techniques}
\subsection{F0 Estimation Techniques}
The most popular F0 estimation algorithms can be categorized as one of
two types:
\begin{itemize}
\item Spectral Techniques
\item Temporal Techniques
\end{itemize}
\subsubsection{Spectral Techniques}
Spectral techniques focus on analysing the spectral content of the signal
by using output from an FFT to perform further processing to determine the
F0 value. Types of technique that can be categorized in this way include:
\begin{itemize}
\item Harmonic Product Spectrum~\parencite[p.8]{smyth2015hps}
\item Cepstral analysis
\item Maximum likelihood
\end{itemize}
\subsubsection{Temporal Techniques}
Temporal techniques attempt to calculate the periodicity of the signal.
This can then be inverted to produce the frequency.
Types of technique that can be categorized in this way include:
\begin{itemize}
\item Autocorrelation~\parencite[p.98]{lerch2012itaca}
\item Zero-crossing~\parencite[p.98]{lerch2012itaca}
\item NCCF (normalized cross correlation function)~\parencite{kasi2015yaapt}
\end{itemize}
\subsection{Current methods for technique combination}
\label{sec:ComMeth}
There is considerably less research into the combination of multiple
algorithms for the improvement of results. Limited research has been
carried out into the effects of training supervised learning algorithms to
pick results based on circumstances.
The ``Yet Another Algorithm for Pitch Tracking'' algorithm attempts to
refine temporal analysis results through the analysis of spectral
information.~\cite{kasi2015yaapt}
Overall there remains a large scope for the type of research proposed.
\section{Methodology}
A detailed analysis of a variety of the most prominent F0 estimation
techniques will be presented, to determine the quantity of methods
needed and which methods will produce the best quality results. Methods
for selecting an algorithm will also require significant further
research. Potential techniques to be explored include:
\begin{itemize}
\item Applying machine learning algorithms in order to
automatically determine the best algorithm as described in
section \ref{sec:ComMeth}~\parencite{bogason2015ffesl}
\item Leveraging information gained from prior feature extraction
to determine aspects such as the signal's noisiness in order to
select the algorithm best suited to this description.
\item Dynamic algorithm parameter adoption to improve the likelihood
of an accurate estimation based on descriptors. Adapting window
size for example.~\parencite{liuni2012aasas}
\end{itemize}
Having determined the optimal set of estimation algorithms and
selection techniques, these will be implemented in python, or
potentially a faster compiled language such as C, to create a tool
capable of creating robust F0 estimations for a range of varying audio
files.
\section{Significance of Research}
This research aims to explore possible improvements to the overall
robustness and general accuracy of pitch detection and thus has
significance in fields such as music and speech analysis and
transformations. By taking a higher level approach to the problem, it
is hoped that the careful combination of algorithms will yield a
superior overall outcome to that of the individual algorithms.
\section{Timeline}
\begin{table}[H]
\centering
\label{my-label}
\begin{tabular}{ll}
Month 1 - 3 & Initial research into methods and combination techniques \\
& as well as the set up of initial framework for code if necessary. \\
Month 3 - 6 & Implementation and testing of individual algorithms. \\
Month 6 - 9 & Implementation of combination methods. \\
Month 9 - 12 & Analysis of results and work on method improvements.
\end{tabular}
\end{table}
\printbibliography
\end{document}
QMUL_Statement_of_purpose/Sam_Perry-Research_masters_proposal.tex
+141
View File
@@ -0,0 +1,141 @@
\documentclass[10pt,letterpaper]{article}
\usepackage{hyperref}
\usepackage{geometry}
\usepackage{enumitem}
\usepackage{multicol}
\newcommand{\tabitem}{~~\llap{\textbullet}~~}
% Fonts
\usepackage[T1]{fontenc}
\usepackage[urw-garamond]{mathdesign}
% Set your name here
\def\name{Samuel Perry}
% The following metadata will show up in the PDF properties
\hypersetup{
colorlinks = true,
urlcolor = black,
pdfauthor = {\name},
pdfkeywords = {DSP, Programmer},
pdftitle = {\name: Curriculum Vitae},
pdfsubject = {Curriculum Vitae},
pdfpagemode = UseNone
}
\geometry{
body={6.5in, 9.0in},
left=1.0in,
top=1.0in
}
% Customize page headers
\pagestyle{myheadings}
\markright{\name}
\thispagestyle{empty}
% Custom section fonts
\usepackage{sectsty}
\sectionfont{\rmfamily\mdseries\Large}
\subsectionfont{\rmfamily\mdseries\itshape\large}
% Other possible font commands include:
% \ttfamily for teletype,
% \sffamily for sans serif,
% \bfseries for bold,
% \scshape for small caps,
% \normalsize, \large, \Large, \LARGE sizes.
% Don't indent paragraphs.
\setlength\parindent{0em}
% Make lists without bullets and compact spacing
\renewenvironment{itemize}{
\begin{list}{}{
\setlength{\leftmargin}{1.5em}
\setlength{\itemsep}{0.25em}
\setlength{\parskip}{0pt}
\setlength{\parsep}{0.25em}
}
}{
\end{list}
}
\setlist[enumerate]{itemsep=0.25em}
\begin{document}
% Place name at left
{\huge\name}
% Alternatively, print name centered and bold:
%\centerline{\huge \bf \name}
\bigskip
\begin{minipage}[t]{0.495\textwidth}
20 Lower Luton Road\\
Wheathampstead\\
Hertfordshire\\
AL4 8QZ\\
\end{minipage}
\begin{minipage}[t]{0.495\textwidth}
Phone: (+44) 7766 521596\\
Email: \href{mailto:samuel.perry89@gmail.com}{samuel.perry89@gmail.com} \\
Linked-in: \\\href{https://uk.linkedin.com/in/sam-perry-04245438}{https://uk.linkedin.com/in/sam-perry-04245438}
\end{minipage}
\section*{\Large Sound and Music Computing MSc \\ \large Statement of Purpose}
The sound and music computing MSc offers a curriculum that is well suited to
continue my studies in the area of audio signal processing. I see the course as
an opportunity to broaden my knowledge of techniques for analysing and
synthesizing sounds digitally. This would build on my current understanding of
these techniques that has been developed over the past four years, during my
time studying at the University of Huddersfield and through working on the
Analysis and Synthesis team in the IRCAM research institute.\\
My time spent at the IRCAM research institute provided me with a valuable
insight into the ways that audio research is carried out and I understand that
the Centre for Digital Music carries out research of a similar nature. For
example I am already familiar with the Sonic Visualiser program which is not
dissimilar to the AudioSculpt software which was used extensively during my
internship at IRCAM. Due to the similarities between the two facilities, I
feel that the style of study on this course would be a logical step forward
from the type of work I encountered at IRCAM.\\
In addition to this I also have a reasonable understanding of audio descriptor
analysis techniques such as pitch and timbre analyses due to research carried
out on my final year project (see
\href{http://pezz89.github.io/pysound/index.html}{http://pezz89.github.io/pysound/index.html}
for details). This would most likely be useful prior knowledge for modules such
as the Music Analysis and Synthesis modules. \\
I would also be interested in other module available such as the machine
learning module, that would give me the opportunity to study a subject I have
basic knowledge of, but have not had the opportunity to explore in detail. I
believe this opportunity would be both interesting and useful for my future
endeavours. I am also keen to develop my programming ability and continue
developing my knowledge of languages such as Python, Matlab and C++. Given the
technical nature of the course, I imagine that my current knowledge of these
languages would be beneficial. \\
Studying on this course would also be an opportunity meet like minded
individuals and develop professional relationships in the industry I wish to
pursue a career in. My internship allowed me to network with a range of
researchers with a variety of specialist subjects and discuss thoughts and
ideas. I found this extremely beneficial to my understanding of this field of
research and would enjoy the oppertunity to network in a similar fashion.\\
I would expect that this course will provide the necessary skills to develop a
career in DSP engineering or provide a basis for further academic research in
these fields. On successful completion of this course I would look to either
further my studies as a PhD candidate or search for a job in commercial DSP or
general programming.\\
Overall I believe that I am a candidate that is well suited to the requirements
of this masters course. Given my previous studies and experience, I am
confident that I have the skill set and attitude required to complete a course
such as this.\\
\newline
Thank you for your consideration.
% Footer
\bigskip
{\small Last updated: \today}
\end{document}