Drum Recorder Research Articles

Comparison of Corecorded Analog and Digital Systems for Characterization of Responses and Uncertainties

Abstract One of the most prominent challenges related to legacy seismic data is determining how these data can be appropriately used in modern research applications. The wide variety of instrumentation used in the analog era, the format of recording on paper wrapped around a helicorder drum, and limited metadata information introduces ambiguities that are not typical of modern digital data. Therefore, techniques must be developed to help characterize uncertainties in legacy data. This article presents an analysis that compares corecorded signals from two instruments—a Trillium Compact or Press-Ewing (PE) seismometer for sensing ground motion and two recording systems: a modern Q330 digitizer or heated-stylus system. Analyses of the recordings in both time and frequency domains indicate time uncertainty on the order of one second, identify a flat response in a 10–60 s band for the PE and drum recorder, and highlight how specific features of scans and paper seismograms (e.g., repeated portions of scans and line thickness) can cause timing jumps or reduced trace amplitude.

A latent rhythm complexity model for attribute-controlled drum pattern generation

Most music listeners have an intuitive understanding of the notion of rhythm complexity. Musicologists and scientists, however, have long sought objective ways to measure and model such a distinctively perceptual attribute of music. Whereas previous research has mainly focused on monophonic patterns, this article presents a novel perceptually-informed rhythm complexity measure specifically designed for polyphonic rhythms, i.e., patterns in which multiple simultaneous voices cooperate toward creating a coherent musical phrase. We focus on drum rhythms relating to the Western musical tradition and validate the proposed measure through a perceptual test where users were asked to rate the complexity of real-life drumming performances. Hence, we propose a latent vector model for rhythm complexity based on a recurrent variational autoencoder tasked with learning the complexity of input samples and embedding it along one latent dimension. Aided by an auxiliary adversarial loss term promoting disentanglement, this effectively regularizes the latent space, thus enabling explicit control over the complexity of newly generated patterns. Trained on a large corpus of MIDI files of polyphonic drum recordings, the proposed method proved capable of generating coherent and realistic samples at the desired complexity value. In our experiments, output and target complexities show a high correlation, and the latent space appears interpretable and continuously navigable. On the one hand, this model can readily contribute to a wide range of creative applications, including, for instance, assisted music composition and automatic music generation. On the other hand, it brings us one step closer toward achieving the ambitious goal of equipping machines with a human-like understanding of perceptual features of music.

Recovering Analog-Tape Seismograms from the 1980 Mount St. Helens Pre-Eruption Period

Abstract Mount St. Helens in Washington State erupted violently (Volcano Explosivity Index = 5) on 18 May 1980. During the previous two months, intense seismic activity at the volcano was recorded by a combination of continuous analog-tape recordings, paper drum recordings, and a recently installed triggered digital event computer system. Because of the technological constraints of the time, the digital data available cover only a little more than 1% of the two-month period. The paper drum records only exist for a few of the seismic stations and are also quite incomplete. However, the analog-tape data from some stations is near complete for almost the whole two months. During the period 2005–2014, these old analog tapes were recovered from storage and digitized to generate standard digital data for archiving at the Incorporated Research Institutions for Seismology Data Management Center. This recovery process was long and complicated but, for the most part, was fairly successful. Although the quality of these recovered data is nowhere near as good as modern digital seismograms, this dataset does provide a near-continuous record of the significant seismic sequence that led up to the major volcanic eruption. It includes the large variety of seismic signals from different types of volcanic earthquakes and harmonic tremor and should be a valuable resource for those studying volcanic seismicity.

Reverse Engineering the Amen Break — Score-Informed Separation and Restoration Applied to Drum Recordings

This work addresses the extraction of high-quality component signals from drum solo recordings (breakbeats) for music production and remixing purposes. Specifically, we employ audio source separation techniques to recover sound events from the drum sound mixture corresponding to the individual drum strokes. Our separation approach is based on an informed variant of non-negative matrix factor deconvolution (NMFD) that has been proposed and applied to drum transcription and separation in earlier works. In this paper, we systematically study the suitability of NMFD and the impact of audio- and score-based side information in the context of drum separation. In the case of imperfect decompositions, we observe different cross-talk artifacts appearing during the attack and the decay segment of the extracted drum sounds. Based on these findings, we propose and evaluate two extensions to the core technique. The first extension is based on applying a cascaded NMFD decomposition while retaining selected side information. The second extension is a time--frequency selective restoration approach using a dictionary of single note drum sounds. For all our experiments, we use a publicly available data set consisting of multitrack drum recordings and corresponding annotations that allows us to evaluate the source separation quality. Using this test set, we show that our proposed methods improve the quality of the component signals.

Quantifying Microtiming Patterning and Variability in Drum Kit Recordings

Human performers introduce temporal variability in their performance of music. The variability consists of both long-range tempo changes and microtiming variability that are note-to-note level deviations from the nominal beat time. In many contexts, microtiming is important for achieving certain preferred characteristics in a performance, such as hang, drive, or groove; but this variability is also, to some extent, stochastic. In this paper, we present a method for quantifying the microtiming variability. First, we transcribed drum performance audio files into empirical data using a very precise onset detection system. Second, we separated the microtiming variability into two components: systematic variability (SV), defined as recurrent temporal patterns, and residual variability (RV), defined as the residual, unexplained temporal deviation. The method was evaluated using computer-performed audio drum tracks and the results show a slight overestimation of the variability magnitude, but proportionally correct ratios between SV and RV. Thereafter two data sets were analyzed: drum performances from a MIDI drum kit and real-life drum performances from professional drum recordings. The results from these data sets show that up to 65 percent of the total micro-timing variability can be explained by recurring and consistent patterns.

Automatic Noise Gate Settings for Drum Recordings Containing Bleed from Secondary Sources

An algorithm is presented which automatically sets the attack, release, threshold, and hold parameters of a noise gate applied to drum recordings which contain bleed from secondary sources. The gain parameter which controls the amount of attenuation applied when the gate is closed is retained, to allow the user to control the strength of the gate. The gate settings are found by minimising the artifacts introduced to the desirable component of the signal, whilst ensuring that the level of bleed is reduced by a certain amount. The algorithm is tested on kick drum recordings which contain bleed from hi-hats, snare drum, cymbals, and tom toms.

Digitization of optical lever instruments

Some classical old-fashioned instruments (such as the Wood-Anderson seismometers), as well as some modern instruments (like the all-quartz made Verbaandert-Melchior tiltmeters), conventionally require an impractical recording system obtained by a photographic drum recorder in a darkroom. Simple electronic equipment (made by readily available, low cost electronics) may help in using such kind of instruments allowing a useful digital recording. This will reduce the time-hour in data acquisition and storage (and the manual error), and will increase the accuracy. The theory of operations and some results obtained using the described equipment are shown.

Local magnitude estimate at Mt. Etna

In order to verify the duration magnitude MD we calculated local magnitude ML values of 288 earthquakes oc- curring from October 2002 to April 2003 at Mt. Etna. The analysis was computed at three digital stations of the permanent seismic network of Istituto Nazionale di Geofisica e Vulcanologia of Catania, using the relationship ML = logA+alog∆−b, where A is maximum half-amplitude of the horizontal component of the seismic record- ing measured in mm and the term «+alog∆−b» takes the place of the term «−logA0» of Richter relationship. In particular, a = 0.15 for ∆<200 km, b =0.16 for ∆<200 km. Duration magnitude MD values, moment magnitude MW values and other local magnitude values were compared. Differences between ML and MD were obtained for the strong seismic swarms occurring on October 27, during the onset of 2002-2003 Mt. Etna eruption, charac- terized by a high earthquake rate, with very strong events (seismogram results clipped in amplitude on drum recorder trace) and high level of volcanic tremor, which not permit us to estimate the duration of the earthquakes correctly. ML and MD relationships were related and therefore a new relationship for MD is proposed. Cumula- tive strain release calculated after the eruption using ML values is about 1.75E+06 J 1/2 higher than the one cal- culated using MD values.

What/when causal expectation modelling applied to audio signals

A causal system to represent a stream of music into musical events, and to generate further expected events, is presented. Starting from an auditory front-end that extracts low-level (i.e. MFCC) and mid-level features such as onsets and beats, an unsupervised clustering process builds and maintains a set of symbols aimed at representing musical stream events using both timbre and time descriptions. The time events are represented using inter-onset intervals relative to the beats. These symbols are then processed by an expectation module using Predictive Partial Match, a multiscale technique based on N-grams. To characterise the ability of the system to generate an expectation that matches both ground truth and system transcription, we introduce several measures that take into account the uncertainty associated with the unsupervised encoding of the musical sequence. The system is evaluated using a subset of the ENST-drums database of annotated drum recordings. We compare three approaches to combine timing (when) and timbre (what) expectation. In our experiments, we show that the induced representation is useful for generating expectation patterns in a causal way.

What/when causal expectation modelling applied to percussive audio

A causal system for representing a musical stream and generating further expected events is presented. Starting from an auditory front‐end which extracts low‐level (e.g., spectral shape, mel frequency cepstral coefficients) and midlevel features such as onsets and beats, an unsupervised categorisation process builds and maintains a set of symbols aimed at representing musical stream events using both timbre and time descriptions. The time events are represented using inter‐onset intervals relative to the beats. These symbols are then processed by an expectation module based on predictive partial match, a multiscale technique derived from N‐grams. The system capacity to generate an expectation that matches its transcription is evaluated using drum recordings from the ENST‐drums database. We show that the MFCC‐based representation leads to a more compact set of symbols and a better match between transcription and expectation. Also, we suggest that the system is sensitive to exposure and illustrate some properties of the expectation entropy while attending percussive audio patterns.

A Real-time Interactive Educational Seismology Exhibit

Modern seismology has rapidly passed the level of many educational displays. While seismologists almost exclusively use digital data, the traditional pen and ink drum recorder remains the mainstay of many educational seismology exhibits in museums. Although a drum recorder retains several major advantages, such as familiarity and visual appeal, it also has several clear disadvantages, both technical and educational. Chief among the technical disadvantages are the poor data quality of many museum displays due to high noise levels and the difficulty and cost of replacing paper, pens, and ink on a routine basis. The use of a drum recorder also gives a misleading and antiquated impression of how modern seismologists work. Another feature of museum displays is that the sensor is usually kept well away from the actual display to reduce seismic noise levels. However, having a sensor near the display is useful for adding interaction to the display. Visitors can create their own seismic signals by jumping, a feature which is especially popular with children. In general, interactivity in museum displays has been shown to increase the attractiveness of the display. Interactivity is believed to promote learning, at least for some learning styles, although it is not always clear exactly what is being learned (Feher, 1990; Falk and Dierking, 1992). In this paper we describe an innovative exhibit that provides interactivity, displays high-quality seismic data, and transmits the data in real-time to a local seismic network. A strong-motion sensor is located under the display, allowing people to jump, stamp feet, and otherwise create seismic waves that appear almost immediately on a computer monitor. During periods of little museum activity, which is at least 14 hours per day, the sensor records useful low-noise data that can be merged with regional seismic network data for scientific analysis. In addition …

Seismicity of the interiors of plates in the Pacific Basin

Historical listings of instrumentally recorded earthquakes raise serious questions about the validity and usefulness of the long‐held presumption that the interiors of ocean plates are aseismic stable masses. Unusual distributions of epicenters within these regions may indicate stress patterns resulting from plate motions, or may indicate nascent subduction zones, ridge systems, or hot spots.Data recorded on ocean‐bottom hydrophones and seismometers in an earlier investigation [Walker and McCreery, 1988] revealed 28 intraplate earthquakes in the deep interior of the Northwestern Pacific Basin that were unreported by the worldwide network of land‐based seismic stations. The data sources included drum recordings of hydrophones near Wake and Enewetok islands in operation for different time intervals from 1963 through 1969, tape recordings from a 9‐element 1500‐km‐long linear array of hydrophones in operation for 2 months in 1981, and tape recordings from September 1982 through 1985 of an 8‐element hydrophone array near Wake Island.

Instrumentation for Simultaneous Recording of Water Level in Digital and Graphical Forms

ABSTRACT THE increasing demand for reliable and detailed water-use data in irrigation has encouraged the use of digital data acquisition systems. An instrumentation system for simultaneous recording of water table heights in digital and graphical forms was developed. The major components of this device were a Stevens Fl recorder, an electronic datalogger, and a float-potentiometer (float-pot) unit. The instrumentation system has worked very well for monitoring water levels at a stream flow station in an irrigation canal, where water level changes rapidly. The Float-Pot unit electronically transferred absolute water level positions to an electronic datalogger in any preset incremental time interval, while mechanically operating the clock-driven pen carriage and chart drum recorder. Data stored in the datalogger were transferred to floppy disk through an RS-232 communication interface and a portable IBM-compatible microcomputer. The raw data were processed, transformed into water levels, and compared with those simultaneously obtained by the chart. The results indicated that the data obtained by the float-pot/electronic datalogger were highly accurate. The proposed sytem has provided an excellent concurrent digital and graphical record of the water level positions with a high level of accuracy.

Modification of a Stevens Type-F Recorder for Simultaneous Recording of Rainfall and Water Level

ABSTRACT A modification made to a Stevens Type-F Water Level Recorder, converting it to a dual-input instrument, enables rainfall and water level to be recorded simultaneously on a common time basis. The conversion uses commercially available components to produce a twin drum recorder. The additional drum is designed to be incrementally rotated by electrical impulses from a tipping bucket rain gauge.

Revised Technique of Photographic Data Acquisition

A previously reported photographic recording device has been modified to accommodate available equipment and conditions. A battery operated system for tripping the shutter of an automatic advance camera, instead of a mechanical linkage, was developed to obtain daily data from a remotely located anemometer. The system includes a rotating drum recorder with an 8-day clock, a 24-volt DC battery power pack, solenoid plunger, capacitor, and microswitch. Cautionary procedures are described for effective operation of the system. A periscope with magnifying lenses was used to locate the camera enclosure and operating mechanism below the plane of the anemometer cups to minimize their effect on the air flow near the instrument. Specific length dimensions of the periscope to provide clear, readable images of the film negative are presented.

Programmed "trendscription." A new approach to electrocardiographic monitoring.

A new approach to monitoring of episodic but frequently recurring arrhythmias has been developed, consisting of condensation of electrocardiographic information on a special drum recorder, a programmer that permits selection of intervals and durations for monitoring, and a telemetry unit for transmission of the electrocardiographic signal. We used this system to manage three selected cases.

Programmed "Trendscription"

A new approach to monitoring of episodic but frequently recurring arrhythmias has been developed, consisting of condensation of electrocardiographic information on a special drum recorder, a programmer that permits selection of intervals and durations for monitoring, and a telemetry unit for transmission of the electrocardiographic signal. We used this system to manage three selected cases. (JAMA232:39-43, 1975)

The seasonal variation in the sound-scattering layer observed at Fletcher's ice island (T-3) with a 12-kc/s echo sounder

FOR NEARLY tWO years, from June 1963 to the present, an echo sounder has been operated continuously at Fletcher's Ice Island (T-3), a drifting research station in the central Arctic Ocean. Diffuse reverberations are frequently seen on the echograms during the summer months. These reverberations are similar in general appearance to the deep scattering layers of other oceans. In non-polar oceans, the deep scattering layer is usually found at depths between 200 and 600 m during the day. At night it rises almost to the surface and weakens or disappears. Two important features distinguish the Arctic Ocean scattering layers from these non-polar scattering layers : (1) the Arctic Ocean scattering layers are relatively shallow, 50-200 m, and (2) the Arctic scattering layers have a pronounced annual, rather than diurnal, cycle. These effects are probably related to light conditions peculiar to polar oceans. First, the light is relatively weak beneath the permanent ice cover. Second, the d a y becomes effectively one year long at these high latitudes so that the Arctic scattering layers are present at moderate depths during the summer light period and then disappear during the winter dark period. In addition to the scattering layers, discrete echoes from shallow depths are frequently recorded. These discrete echoes occur throughout the year but are particularly frequent during the winter. These echoes take a hyperbolic shape indicating relative movement between the ice station and the reflector. Presumably they are caused by nekton such as fish or seals. This is believed to be the first observation of sound scattering layers in the central Arctic Ocean. An examination by DIETZ and SHUMWAY (1961) of echograms taken on the nuclear submarines Nautilus and Skate during their Arctic Ocean cruises revealed no deep scattering layers or discrete echoes. However, the submarines probably cruised within or below the scattering layer. The echo sounder at the ice station was better placed for detecting a shallow scattering layer. T-3 drifted along the track shown in Fig. 1 during the period of echo sounding measurements. The Precision Depth Recorder system consisted of a standard Edo transducer, a spark-type drum recorder and a sonar transceiver manufactured by T. H. Gifft and Associates. The system operated with an electrical power output of 800W at a frequency of 12 kc/s. The outgoing ping was 80/zsec (l cycle) in length and was repeated once each second. The drum rotated once each second

Recording-Reproducing System Using a Dielectric Storage Medium

A method of short-term storage of an electrical signal as a surface charge on a dielectric drum is described. An ionization recording head is used to produce a localized surface charge on a rotating plastic-coated drum, and a capacitive pickup probe is used to reproduce the signal. Basic design considerations and operating characteristics of the recording and reproducing heads are presented along with over-all response curves for an experimental recorder. A comparison with magnetic drum recording illustrates the advantages of the dielectric recorder to be found in the simplicity of probe and recording head design and the appreciably lower machining precision required. The finite life-time of the recording head appears to be long enough to be unimportant for most applications. The symmetrical recording and erasing characteristics of the recording head which are linearly proportional to potential difference between the surface charge and the applied recording voltage permit cumulative recording to be carried out for hundreds of drum revolution periods. This property of the dielectric recorder makes it uniquely suited for multiple sweep integration applications. (This work was sponsored by Bureau of Ships contract NObsr 43356 NE-120221-5.)

A flux-sensitive reproducing head for magnetic recording systems

A description is given of a new type of flux-sensitive reproducing head for use on magnetic tape or drum recording systems. The head is compact and relatively simple in operation and differs from conventional reproducing heads in that its efficiency is fundamentally independent of signal frequency over the working range and is completely independent of the speed of the recording medium on reproduction. It can, in fact, be used to make measurements of recorded flux at points along the length of a stationary recorded medium. The head may be of value in certain specialized applications of magnetic recording technique, such as those dealing with very-low-frequency signals or the examination of transient phenomena.