Hi, I'm Joren. Welcome to my website. I'm a research software engineer in the field of Music Informatics and Digital Humanities. Here you can find a record of my research and projects I have been working on. Learn more »
A new version of Tarsos was uploaded today and it contains an exciting (at least my kind of exciting) new feature. It is capable of real-time pitch analysis and tone scale construction. A video should make its use clear:
The immediate feedback is practical for educational purposes: it makes rather vague things like quarter tones or (uncommon) pitch intervals in general quite tangible. It could be used by singers or string players to explore microtonality or to improve their technique. Another use case is ethnomusicologic field-work: if you would want to research Kora tuning (an African harp) Tarsos could be a practical tool for real-time analysis.
Naar jaarlijkse gewoonte wordt er in het Orpheus instituut de Dag van het Artistiek onderzoek georganiseerd. Hieronder volgt een tekstje over het onderzoeksproject rond Tarsos dat in het jaarboek komt. Het jaarboek is een boekje met daarin een overzicht van artistieke onderzoeksprojecten aan Vlaamse instituten. Het wordt gepubliceerd naar aanleiding van de eerder aangehaalde “Dag van het Artistiek Onderzoek”.
Het doel van dit onderzoeksproject is het ontwikkelen van een methode om een cultuuronafhankelijke kijk op muzikale parameters te verkrijgen. Meer concreet worden er technieken aangewend uit Music Information Retrieval om toonhoogte, tempo en timbre te bestuderen. Aanpassing van bestaande, meestal westers georiënteerde, MIR-methodes moet leiden tot een gestructureerde documentatie van verschillende klankkleuren, toonschalen, metrische verhoudingen en muzikale vormen. Die beschrijving kan dienen als inspiratie voor de ontwikkeling van een artistieke compsitionele taal of kan gebruikt worden als bronmateriaal voor wetenschappelijk onderzoek rond ethnische muziek. Bijvoorbeeld om (de eventuele\
teloorgang van) de eigenheid van orale muziekculturen objectief aan te tonen.
In de eerste fase van het onderzoek ligt de focus van het onderzoek op één van de meer tastbare parameters: toonhoogte. In etnische muziek is het gebruik van toonhoogte vaak radicaal anders dan westerse muziek die meestal gebaseerd is op de onderverdeling van een octaaf in twaalf gelijke delen. Om toonladders uit\
muziek te extraheren en weer te geven werd het software platform Tarsos ontwikkeld. Met Tarsos is het mogelijk om automatische toonladderanlyse uit te voeren op een grote dataset of om manueel een gedetailleerde analyse te verkrijgen van enkele muziekstukken. De cultuuronafhankelijke analysemethode waarvan Tarsos gebruik maakt kan even goed toegepast worden op Indonesische, Westerse of Afrikaanse muziek.
Onze bedoeling is om Tarsos te gebruiken om evoluties in toonladdergebruik te ontdekken in de enorme dataset van het Koninklijk Museum voor Midden-Afrika. Is toonladderdiversiteit in Afrika aan het wegkwijnen onder invloed van Westerse muziek? Zijn er specifieke kenmerken te vinden over eventueel 'uitgestorven' muziekculturen? Dit zijn vragen die kaderen in het overkoepelende onderzoeksproject van Olmo Cornelis en waar we met behulp van Tarsos een antwoord op proberen te vinden.
Later krijgen de twee overige muzikale parameters, tempo en timbre, een gelijkaardige behandeling. In de laatste fase van dit toch wel ambitieuze onderzoekproject wordt de relatie tussen de parameters onderzocht.
At the workshop I had an interesting meeting with Dan Tidhar. He researches harpsichord temperament estimation at QMUL (Queen Mary University of London). Together they created the Tempest web service where you can upload harpsichord audio and let the system guess the temperament. The process is described in the paper “High precision frequency estimation for harpsichord tuning classification”. Although Tarsos was not officially part of the programme I hijacked the poster sessions to show a live demo of Tarsos with Dan’s dataset.
Another interesting talk was about 2032, a tunable synthesizer with definable Harmonics. It elaborates on the ideas of Sethares about tone scales .
Monday, I’ll give a small presentation about Latex and Version Control for the research team at the University College Gent, Faculty of Music. The idea is to give a pragmatic overview of working with Latex and version control. The “presentation about Latex & Version control (in Dutch)”:[latex_svn_presentatie.pdf] can be downloaded. The presentation itself is created using Latex and “the source of the presentation”:[latex_svn_presentatie.zip] is also available. A good description of Latex can be found here:
LaTeX (pronounced “latech”) is a document preparation system for high-quality typesetting based on, and succeeding TeX formatting. It is a very popular format in academia, as it allows advanced document formatting capabilities not found in other common document formatting systems. Some of these capabilities include table figure notations, bibliography formatting (see BibTeX), and an advanced macro language.
Eclipse: an IDE (Integrated Development Environment) which can be used as Latex editor with version control by installing the TeXclipse and Subclipse plugins.
This post contains links to genuinely useful software to do signal based audio analysis.
Sonic Visualizer: As its name suggests Sonic Visualizer contains a lot different visualisations for audio. It can be used for analysis (pitch,beat,chroma,…) with VAMP-plugins. To quote “The aim of Sonic Visualiser is to be the first program you reach for when want to study a musical recording rather than simply listen to it”. It is the swiss army knife of audio analysis.
BeatRoot is designed specifically for one goal: beat tracking. It can be used for e.g. comparing tempi of different performances of the same piece or to track tempo deviation within one piece.
Tartini is capable to do real-time pitch analysis of sound. You can e.g. play into a microphone with a violin and see the harmonics you produce and adapt you playing style based on visual feedback. It also contains a pitch deviation measuring apparatus to analyse vibrato.
Tarsos is software for tone scale analysis. It is useful to extract tone scales from audio. Different tuning systems can be seen, extracted and compared. It also contains the ability to play along with the original song with a tuned midi keyboard .
Melodic Match is a different beast. It does not work on signal level but processes symbolic audio. More to the point it searches through MusicXML files - which can be created from MIDI-files. See its website for use cases. Melodic Match is only available for Windows.
During a lecture at the University College Gent, Faculty of Music these tools were presented with some examples. “The slides (Slides in Dutch)”:[presentatie.pdf] and “a zip-file with audio samples, slides and software”:[Tarsos-presentatie.zip] are available for reference. Most of the time was given to Tarsos, the software we developed.
Olmo Cornelis also gave a lecture about his own research and how Tarsos fits in the bigger picture. “His presentation (Slides in English)”:[lezing_Olmo_Cornelis_zonder_audio.ppt] and “the presentation with audio”:[lezing_Olmo_Cornelis_met_audio.zip] are also available here.
There is more to Tarsos then meets te eye. The graphical user interface only exposes some functionality; the API (Application Programmer Interface) exposes all of Tarsos’ capabilities.
Tarsos is programmed in Java so the API is accessible trough Java and other programming languages targeting the JVM (Java Virtual Machine) like JRuby, Scala and Groovy. The following examples use the Groovy programming language because I find it the most aesthetically pleasing with regards to interoperability and it gets the job done without getting in your way.
To run the examples a copy of the Tarsos JAR-file needs to be added to the Classpath and the Groovy runtime must be installed correctly. I’ll leave this as an exercise for the reader: godspeed to you, brave soul. Quick protip: placing a copy of the jar in the extensions directory seems to work best, e.g. see important java directories on mac OS X.
The first example extracts pitch class histograms from a bunch of files and saves them as EPS (Encapsulated PostScript)-files. It iterates a directory recursively and handles each file that matches a given regular expression. In this example the regular expression matches all WAV-files. Batch processing is one of those things scripting is ideal for, doing the same thing with the user interface would be tedious or even mind-numbingly boring, not groovy at all indeed.
FileUtils.glob(dir,”.*.wav”,true).each { file ->\
audioFile = new AudioFile(file)\
pitchDetector = PitchDetectionMode.TARSOS_YIN.getPitchDetector(audioFile)\
pitchDetector.executePitchDetection()\
//get some annotations\
annotations = pitchDetector.getAnnotations()\
//create an ambitus and tone scale histogram\
ambitusHistogram = Annotation.ambitusHistogram(annotations)\
toneScaleHisto = ambitusHistogram.toneScaleHistogram()\
//plot a smoothed version of the histogram\
p = new SimplePlot()\
p.addData 0, toneScaleHisto.gaussianSmooth(0.2)\
p.save FileUtils.basename( file) + “.eps”\
}\
```
The second example uses functionality that is currently only available trough the API. It takes a MIDI-file and synthesizes it to a wave file using an arbitrary scale. In this case 10-TET. The heavy-work is done by the Gervill synthesizer. The resulting file is available for download, micro—macro?—tonal Bach is great: “BWV 1013 in 10-TET”:[BWV_1013_10-TET.mp3]. The result of “an analysis with Tarsos on the synthesized audio”:[120.png] clearly shows an interval of 120 cents with some deviations.
midiFile = new File(“BWV_1013.mid”)\
outFile = new File(“out.wav”)
tuning = [0,120,240,360,480,600,720,840,960,1080] as double []
MidiToWavRenderer renderer\
renderer = new MidiToWavRenderer()\
renderer.setTuning(tuning)\
renderer.createWavFile(midiFile, outFile)\
```
An extended version of this second example script could be used to generate a dataset with audio and corresponding tone scale information on the fly. The dataset could then be used as a baseline.
The API is not yet well documented and is still in flux or more correctly: superflux. Note to self: I will provide documentation and a number of useful examples when the dust settles down. I’m not even sure if I will stick with Groovy. Scala has a nice Lispy feel to it and seems more developed. Groovy has a less steep learning curve, especially if you have some experience with Ruby. JRuby is also nice but the interoperability with legacy Java looks like an ugly hack.
Yesterday Tarsos was publicly presented at the symposium Perspectives for Computational Musicology in Amsterdam. The first public presentation of Tarsos, excluding this website. The symposium was organized by the Meertens Institute on the occasion of Peter van Kranenburg’s PhD defense.
The presentation included a live demo of a daily build of Tarsos (a Friday evening build) which worked, surprisingly, without hiccups. The presentation was done by Olmo Cornelis. This was the small introduction:
Tarsos - a Platform for Pitch Analysis of Ethnic Music \
Ethnic music is a vulnerable cultural heritage that has received only recently more attention within the Music Information Retrieval community. However, access to ethnic music remains problematic, as this music does not always correspond to the Western concepts of music and metadata that underlie the currently available content-based methods. During this lecture, we like to present our current research on pitch analysis of African music. TARSOS, a platform for analysis, will be presented as a powerful tool that can describe and compare scales with great detail.
Drag and drop works for scala tone scale files and different kinds of audio files. Audiofiles are transcoded automagically using an embedded ffmpeg binary which is platform dependend. It works on linux and windows, on other platforms only WAV files are supported.
Some of the current features:
Scala file extraction from audio
Real time pitch tracking
Real time pitch class histogram visualization
Alignment of pitch intervals with histogram using mouse dragging
Tarsos can be used to render MIDI files to audio (WAV) files using arbitrary tone scales. This functionallity can be used to (automatically) verify tone scale extraction from audio files. Since I could not find a dataset with audio and corresponding tone scales creating one using MIDI seemed a good idea.
MIDI files can be found in spades, tone scales on the other hand are harder to find. Luckily there is one massive source, the Scala Tone Scale Archive: A large collection of over 3700 tone scales.
Using Scala tone scale files and a midi files a Tone Scale - Audio dataset can be generated. The quality of the audio depends on the (software) synthesizer and the SoundFont used. Tarsos currently uses the Gervill synthesizer. Gervill is a pure Java software synthesizer with support for 24bit SoundFonts and the MIDI tuning standard.\
How To Render MIDI Using Arbitrary Tone Scales with Tarsos
A recent version of the JRE (Java Runtime Environment) needs to be installed on your system if you want to use Tarsos. Tarsos itself can be downloaded in the form of the “Tarsos JAR Package”:[tarsos.jar].
Currently Tarsos has a Command Line Interface. An example with the files you can find attached:
The result of this command should yield an audio file that sounds like “the cello suites of bach in a nonsensical tone scale with steps of 120 cents”:[bach_BWV_1007_120.mp3]. Executing tone scale extraction on the generated audo yields the expected result. In the pich class histogram every 120 cents a peak can be found.
To summarize: by rendering audio with MIDI and Scala tone scale files a dataset with tone scale - audio information can be generated and tone scale extraction algorithms can be tested on the fly.
This method also has some limitations. Because audio is rendered there is no (background) noise, no fluctuations in pitch and timbre,… all of which are present in recorded audio. So testing testing tone scale extraction algorithms on recorded audio remains advised.
A new version of Tarsos was uploaded today and it contains an exciting (at least my kind of exciting) new feature. It is capable of real-time pitch analysis and tone scale construction. A video should make its use clear:
Tarsos, featured, Computational ethnomusicology, and HoGent
In de eerste fase van het onderzoek ligt de focus van het onderzoek op één van de meer tastbare parameters: toonhoogte. In etnische muziek is het gebruik van toonhoogte vaak radicaal anders dan westerse muziek die meestal gebaseerd is op de onderverdeling van een octaaf in twaalf gelijke delen. Om toonladders uit\
muziek te extraheren en weer te geven werd het software platform Tarsos ontwikkeld. Met Tarsos is het mogelijk om automatische toonladderanlyse uit te voeren op een grote dataset of om manueel een gedetailleerde analyse te verkrijgen van enkele muziekstukken. De cultuuronafhankelijke analysemethode waarvan Tarsos gebruik maakt kan even goed toegepast worden op Indonesische, Westerse of Afrikaanse muziek.
Onze bedoeling is om Tarsos te gebruiken om evoluties in toonladdergebruik te ontdekken in de enorme dataset van het Koninklijk Museum voor Midden-Afrika. Is toonladderdiversiteit in Afrika aan het wegkwijnen onder invloed van Westerse muziek? Zijn er specifieke kenmerken te vinden over eventueel 'uitgestorven' muziekculturen? Dit zijn vragen die kaderen in het overkoepelende onderzoeksproject van Olmo Cornelis en waar we met behulp van Tarsos een antwoord op proberen te vinden.
Later krijgen de twee overige muzikale parameters, tempo en timbre, een gelijkaardige behandeling. In de laatste fase van dit toch wel ambitieuze onderzoekproject wordt de relatie tussen de parameters onderzocht.
tarsos_orpheus_jaarboek.txt and tarsos_orpheus_jaarboek.pdf
Yesterday Tarsos was publicly presented at the symposium 
