I've recently moved onto some new music information retrieval projects. I moved on from Humtap (wishing them ongoing success!) to MuseAmi in June but unfortunately they made a big strategic pivot in August and downsized the company by a third, myself included. Sad of course, but ultimately start-ups sometimes need to make tough decisions.
The project I've been working on for the last eight months has now released our first public release. Humtap is a mobile app for collaborative music making. Users collaborate by each recording hums and taps into their mobile phones. These audio recordings are transcribed and become musical avatars. A user can then create a piece of music (currently of the electronica genre) by combining their hum or tap with another user's tap or hum, or with their own tap or hum. Finally, users can also create a new piece of music by combining two other user's hum and tap. The app combines the transcriptions in a musical way and selects the electronic instrumentation.
Since March, I am now independently contracting to a number of companies doing music and audio DSP and machine learning. This is a great opportunity for me to work with and advise a number of really great companies, particularly small startups, on commercial applications of audio signal processing, machine learning, information retrieval, cloud infrastructure. You can check my biography for more details.
I've had an Apple Airport Express 1st Generation, 802.11g model A1084 since new, c. 2004. This has long been superseded by newer versions, and for sometime it was just doing duty for me as a USB print server, not as a router. However, it seems that there is a bug introduced around v6.2 of the firmware that would cause it to go offline when configured to "join wireless network". Restarting the AE would allow it to run, but it would soon drop off the net. It became particularly troubling as it would become unresponsive almost as soon as it was configured, barely even allowing a single print job to be sent. It's not clear what the cause is or where exactly the bug may lie.
I'm woefully late in pointing this out, but there is now a video done by Matt Hines and Jay Leboeuf explaining MediaMined.
Proceedings of the 12th International Conference on Music Perception and Cognition, page 943, Thessaloniki, Greece, July 2012. ICMPC/ESCOM. (abstract).
A software system, MediaMined, is described for the efficient analysis and classification of auditory signals. This system has been applied to the tasks of musical instrument identification, classifying musical genre, distinguishing between music and speech, and detection of the gender of human speakers. For each of these tasks, the same algorithm is applied, consisting of low-level signal analysis, statistical processing and perceptual modeling for feature extraction, and then supervised learning of sound classes. Given a ground truth dataset of audio examples, textual descriptive classification labels are then produced. Such labels are suitable for use in automating content interpretation (auditioning) and content retrieval, mixing and signal processing. A multidimensional feature vector is calculated from statistical and perceptual processing of low level signal analysis in the spectral and temporal domains. Machine learning techniques such as support vector machines are applied to produce classification labels given a selected taxonomy. The system is evaluated on large annotated ground truth datasets (n > 30000) and demonstrates success rates (F-measures) greater than 70% correct retrieval, depending on the task. Issues arising from labeling and balancing training sets are discussed. The performance of classification of audio using machine learning methods demonstrates the relative contribution of bottom-up signal derived features and data oriented classification processes to human cognition. Such demonstrations then sharpen the question as to the contribution of top-down, expectation based processes in human auditory cognition.
If you are a budding music or audio engineering undergraduate student, iZotope is hiring paid interns. The work mostly consists of auditioning our systems.
The company I have been working with for about 18 months now, Imagine Research has now been acquired by iZotope Inc. based in Boston. This is a great opportunity for us to continue to develop our MediaMined technology, to broaden it's reach and incorporate it into iZotope's future products. For my local friends, I'll continue to be based in NYC.