Listen launched the first commercial software and soundcard based audio test system in 1995, and has a 25 year history of firsts in audio measurement. We are immensely proud of our many algorithms, measurement techniques and products that we have launched over the years. Here are the highlights.
Listen Celebrates 25th Anniversary
Listen celebrates its 25th anniversary, still under the leadership of its founder, Steve Temme. We would like to thank all our wonderful employees, industry colleagues and SoundCheck users worldwide, who have made this possible. We hope that the world with be in a position where we can celebrate properly sometime in the not too distant future.
AmpConnect 621 is a rugged 6-in, 2 out high resolution audio interface specifically developed for multichannel audio test applications such as automotive audio, sound power measurements, and measuring devices with multiple speakers such as smart phones or sound bars. In addition to 6 inputs and 2 outputs, AmpConnect 621 offers amplifier output, microphone power, full TEDS calibration support, impedance measurement and digital I/O in a single USB-connected unit.
Designed for seamless integration with SoundCheck, it is controlled via the SoundCheck software only. Setup is fully automatic, and there are no front panel buttons, eliminating the possibility of accidental changes to the settings and making it ideal for production applications.
New Auto-Delay+ Algorithm
A new Auto Delay+ algorithm offers increased accuracy for open loop measurements, especially in cases where the delay is longer than 1 second. It is particularly useful for measuring devices where test signals are sent via the cloud, such as smart speakers, automotive audio, hearables, etc.
TEDS Calibration Support
TEDS Calibration Support (with compatible Listen hardware) enables automatic identification, configuration and calibration of TEDS microphones and accelerometers, saving time on initial hardware setup and whenever hardware is changed.
Read Multi-Channel WAV files
Testing multichannel devices such as microphone or speaker arrays is much easier and faster as SoundCheck now reads multichannel WAV files from the memory list, signal generator and stimulus steps; there is no need to de-interleave the files in a separate utility before importing into SoundCheck. When opening a multichannel WAV file in the Memory list, each WAV file channel is named using the file name and channel number, and all channels are automatically grouped together.
Background Noise Simulation
Listen offers a test sequence to calibrate a 4.1 speaker array to conform with the ETSI ES 202 396-1 standard. This is considerably less expensive than stand-alone background noise systems, and has the added advantage that it can be integrated into test sequences.
Level and Frequency based triggering is introduced in SoundCheck, enabling acquisition to be triggered from an external source using a pilot tone at the beginning of a test sweep. Just a year later, this is upgraded with the addition of a Level and Cross-Correlation Trigger, using a log chirp signal instead of a pure frequency tone. This enhances performance for open loop measurements as it is more robust and less susceptible to false triggers.
Sweep Equalization for Minimized transients
Stepped sine amplitude and frequency sweeps can be equalized for a smooth transition between steps. This minimizes the transient response in the device under test, resulting in faster measurements.
New Strip Chart Recorder
A new strip chart recorder expands the capabilities of the the multimeter, distortion analyzer and frequency counter with the ability to plot measurements over time – the electronic equivalent of connecting a paper chart recorder to a hardware instrument. It plots continuously, or for a predefined amount of time, and can record instantaneous results or repeating averages. This feature is valuable for environmental and reliability testing and in any situation where the behavior of a device over time must be characterized. Results from the strip chart recorder can be saved to the memory list and used inside a sequence just like any other virtual instrument.
New Real time Distortion Analyzer
A new real time distortion analyzer enables distortion, including THD and THD+N, THD and THD+N residual level, and SINAD, to be quickly viewed in real time without having to configure a sequence. A, B, and C weighting filters along with user-defined arbitrary weighting functions can be applied. Data from the distortion analyzer can also be saved and, with the strip chart recorder module installed, viewed over time.
New THD+N Algorithm
A new, AES-17 compliant, THD+N Analysis Algorithm is more accurate and significantly faster than other audio test systems. The algorithm also includes high pass and low pass filters for controlling the measurement bandwidth and for filtering noise in electronic measurements.
Powerful TCP/IP Control
SoundCheck now offers powerful external control with TCP/IP (in addition to Active X). This enables SoundCheck to be controlled via any programming language, on any operating system, either locally or through a network. This powerful and expandable command format offers huge benefits for factory automation as a single computer can control multiple SoundCheck systems.
This was further expanded in 2020 with the new sequence parameter feature, which allows the user to use TCP/IP commands to pass test configuration data into the memory list from external programs. By externally storing parameters such as limits, test levels, and test signals, a single sequence can be used for multiple products, simplifying the number of sequences that a large organization needs to maintain, and reducing test configuration time. It is also useful for applications where a sequence needs to be run many times with different parameters, for example, testing voice recognition with a range of voices or test levels.
Autoranging Optimizes Dynamic Range of Test
SoundCheck’s new Auto-ranging feature automatically adjusts the microphone preamplifier gain in order to optimize the dynamic range of the test for optimum signal to noise ratio. With this enabled, SoundCheck monitors the digital headroom of the audio interface and increases or decreases the preamp gain as necessary. When a change to the gain is made the acquisition step runs again with the new settings.
SoundCheck for Mac
SoundCheck is the first commercially available large-scale audio test solution to run under the Mac operating system. Sequences, steps, and data files are cross compatible between the Windows and Mac versions, providing versatility when collaborating with other SoundCheck users, and Core audio driver support allows the same hardware to be used on both operating systems.
Listen’s AudioConnect is an extremely compact and cost-effective 2 channel Audio Interface. Designed specifically for portable audio measurement applications, it is packed with additional features to simplify audio test configurations, including SCM microphone power with 0dB/20dB gain, a headphone amplifier and a USB hub for daisy-chaining of USB components.
Listen’s new entry level SoundCheck package consists of an audio interface packaged with a simplified version of the SoundCheck software. SoundCheck ONE includes sequence templates that are used to generate product-specific sequences by selecting parameters such as the stimulus signal, characteristics to be measured, frequency range, level and limits. This makes the system very simple to use, in addition to being extremely cost-effective. It can be upgraded to a higher level SoundCheck package at any time.
SoundConnect 2 is a compact and rugged USB controlled microphone and accelerometer power supply and conditioning amplifier. It powers either one or two microphones/accelerometers, so it is a great choice for audio and vibration test applications where two transducers need to be powered simultaneously, for example headphone testing, stereo measurements, or anywhere both a reference microphone and input microphone require power. It offers both DC and IEPE power, and optional LEMO connectors for applications requiring 200V polarization. Naturally it is fully controllable from within SoundCheck.
AudioConnect 4×4 is a high accuracy 4-in, 4-out audio interface specifically for audio test and measurement. Controls are simple – a single USB connection and a power indicator – all other typical soundcard controls have been eliminated to make operation and calibration simple and error-free.
SC AMP Amplifer
SC AmpTM, the first amplifier designed specifically for audio testing, incorporates several features unique to the needs of audio test and measurement, and unnecessary features found in regular audio amplifiers (e.g. volume knobs) have been eliminated to keep operation simple and error-proof. SC Amp’s integrated impedance measurement and DC output uses Listen’s unique patented high accuracy single channel impedance measurement method to offer dual channel accuracy using just a single channel. SC Amp is fully controlled via SoundCheck, and will even work with other brands of test systems.
Bluetooth and MEMS Interfaces
Listen partners with Portland Tool and Die to offer innovative interfaces that enable products such as Bluetooth and MEMS microphones to be tested. These can be fully controlled via SoundCheck, expanding the system’s test capabilities.
Batch processing functionality now enables the user to carry out the same operation on multiple data sets simultaneously. This is useful for examining large quantities of production data and tweaking limits to see the effect on yield, or when it is necessary to change the resolution of many curves.
Time Domain Waveform Filters
Time Domain Waveform Filters allow a curve from the memory list to be convolved with a waveform in the memory list to create a new waveform that has its spectral content shaped by the selected curve. There is also an option for minimum phase and inverting the curve. This may be used, for example, for measuring the A-weighted peak acoustic pressure of a waveform. The A-weighting curve is first applied to the acoustic waveform via the waveform filtering post processing step, and then the peak value of the resulting waveform is measured. This method is used in the IEEE 269 and TIA 920 telephony standards.
Listen is the first to integrate Zwicker loudness into an electroacoustics test package. Zwicker loudness calculates the overall perceived loudness of a sound using a psycho-acoustic model which takes into account the nonlinearity of the human ear to sound at different frequencies and levels. It provides the capability to measure the perceived loudness of complex sounds, e.g. telephone ringtones.
Perceptual Rub & Buzz
Listen is the first company to offer perceptual Rub & Buzz measurements to detect only those Rub & Buzz faults that would be heard by the human ear. The CLEAR (Cepstral Loudness Enhanced Algorithm for Rub & Buzz) algorithm offers true perceptual Rub & Buzz analysis using a simplified auditory perceptual model to measure the loudness of Rub & Buzz distortion in phons rather than the more traditional dB SPL and % distortion units. This better identifies whether distortion due to manufacturing defects can be heard by the listener than conventional measurements. In addition to a result which corresponds more accurately to the human ear, this new test method is also less sensitive to transient background noises than traditional methods, and it is much simpler to set limits.
AmpConnect ISC is a complete, self-contained audio measurement interface which provides a single-box solution for measuring speakers, headphones and microphones. It is compact, simple to set up, and lowers system cost, making it equally suitable for laboratory and production line use. In addition to the audio interface, it provides microphone power, drives speakers and headphones, and includes impedance measurement circuits, replacing multiple hardware devices. Further features include automated control of signal routing, input gain and instrument configuration. It connects to a computer via a single USB cable and is fully integrated with SoundCheck.
Wavefile Streaming with Real Time Equalization
The signal generator now offers wavefile streaming with real time equalization. This removes memory limitations on the length of test signals, which means that longer test signals such as those required for analysis of speech and music can easily be accommodated.
Listen is the first to combine CSD, STFT, Wigner-Ville and Wavelet transforms into an easy-to-use software package. SoundMap (now integrated as a SoundCheck module) is ideal for impulse response analysis and detection of loose particles and Rub & Buzz in loudspeakers. It is also valuable for identification of transient effects such as drop outs in digital devices including VoIP and Bluetooth headsets. It offers greater flexibility than conventional ‘waterfall plot’ software, and more accurate and psychoacoustically significant analysis options. In addition to analyzing SoundCheck data, it reads any WAV, ASCII or MLSSA TIM file for analysis of data collected on other platforms.
Open Loop Testing
Listen was the first to introduce Open Loop testing – testing a device where the stimulus and response are asynchronous. This method was originally developed to test MP3 players, and involved loading a test signal onto the device, then using a triggered acquisition step to capture the response. These days, open loop testing is most widely used for testing voice-activated devices such as smart speakers, automotive audio, hearables and more. SoundCheck’s algorithms have evolved over the years and now include sophisticated frequency and chirp triggers, advanced auto-delay algorithms and more.
Frequency Shift and Resampling
Listen introduces frequency shift and resampling, which enables electronics with asynchronous clocks, such as MP3 players, to be tested. In recent years, these same algorithms have proved invaluable for open loop testing of smartphones, smart speakers and other voice activated devices.
Listen is the first to offer the ability to play a conditioning signal right before the excitation signal. This enables a speaker to be tested in steady state conditions, for example by playing pink noise first to warm up the speaker, or in telephony applications it enables a signal to be used to trigger a voice activity detector before beginning recording. The conditioning signal is excluded from the response analysis. In more recent years, this same stimulus flexibility has allowed the use of sophisticated stimuli for open loop test sequences for smart and voice-activated devices.
Full Multichannel Acquisition
Full multichannel acquisition enables multi-channel, or even multiple multi-channel Soundcards to be used with SoundCheck; the number of available channels is limited only by the computer’s speed and memory. Play and record can even be implemented on different devices, for example, allowing a high resolution NI data acquisition card to be used in conjunction with a soundcard and a Bluetooth interface. Extra channels can be added in pairs, these days, up to a maximum of 128, enabling a system to evolve with changing test needs. Many systems that were sold as basic 2-channel systems have been upgraded to now offer 8 or more channels as measurement requirements have changed!
Non-Coherent Distortion with Music
Listen was the first test and measurement system to offer an algorithm for the measurement of non-coherent distortion using music as the test signal. This algorithm uses the transfer function to examine the part of the response that is not coherent with the stimulus, enabling total distortion versus frequency to be measured under real world conditions using signals such as music or speech.
Loose Particles Algorithm
Listen is the first audio measurement software to include a loose particle/transient detection algorithm. This algorithm uses time envelope analysis to catch random transient events such as loose particles in the time domain and quantifies their level over time. This analysis enables rapid and accurate production line detection of particles trapped in a loudspeaker’s motor/voice coil area, a manufacturing defect hard to catch with traditional rub & buzz measurement techniques.
DC Connect Precision DC Power Supply and Current Monitor
DC Connect is USB controlled precision DC power supply and current monitor. It enables high resolution dynamic measurements of voltage and current consumption in portable electronic devices such as hearables, hearing aids, microphones, cell phones, and other battery powered electronic devices. Equally suitable for R&D and production line use, full software control, enables the output level to be changed rapidly during a test to enable measurement of current consumption while varying frequency and amplitude of the test signal.
Save to Database
SoundCheck exports test results to Word, Excel and Databases. Although we take these things for granted these days, back when we were competing against human listeners in production line test applications, this was a pretty revolutionary feature! Naturally, this functionality has seen continuous upgrading over the years. Now, for laboratory users, in addition to exporting to Word and Excel for report-writing, we can also automatically save to Matlab for additional analysis. Stimuli and responses can be saved as WAV files with the test data for full traceability and reference of test results. For production users, the enhanced database option offers extremely fast and space-efficient storage of large volumes of data. This saves the data to a binary format; SQL and Python tools are provided to facilitate data extraction so that detailed analysis and graphical representation may be implemented in your Business Intelligence (BI) system of choice (e.g. Tableau, Qlik,etc.). SoundCheck data can also be imported, via a converter, into WATS. Additionally, data collection can be integrated as part of a larger factory automation system (e.g. TestStand) or even saved to a custom database.
Log TSR - First Commercial Implementation of the Farina Sweep
The Log TSR measurement method (also known as ‘Farina Sweep’ or ‘Log Chirp’) was already under development in SoundCheck when Angelo Farina presented his landmark paper on the subject, so within weeks of this method reaching a wider audience, SoundCheck had the first commercial implementation Log TSR measurement methods – a technique still extensively used today for making fast measurements. This continuous log sine sweep from low to high covers every frequency in the chosen range using the same time (and energy) for every octave, which achieves a good signal to noise ratio across all frequencies. Since the sweep rate is slow at low frequencies but increases with frequency, distortion is accurately measured at low frequencies while retaining a fast test speed and high throughput. This makes it a common choice for rapid loudspeaker production line measurements of fundamental response and distortion. Used together with a Time Selective Response algorithm, simulated free field measurements can be made. This technique enables anechoic chamber measurements to be made in a normal room with no loss of accuracy.
SoundConnect Microphone Power Supply
SoundConnect, Listen’s first hardware product, is a rugged and affordable microphone power supply. This was followed in 2000 by the Listen SCM, a rugged, low-cost ¼” measurement microphone specially designed to survive harsh factory environments. These two products significantly reduce the cost of a complete audio test system, making it far more affordable for production testing. The product has been so successful that it is still sold today!
Listen is the first test and measurement system to feature virtual instruments – software equivalents of a signal generator, multimeter, real time analyzer, oscilloscope and spectrum analyzer. These virtual instruments can be programmed and used within test sequences, or used as stand-alone instruments. Multiple instances of the same instrument can even be opened simultaneously, enabling real time live data to be viewed in multi-channel applications. This naturally makes for an extremely compact and flexible setup for the laboratory, traveling, and these days, working from home. In more recent years, the instrument panel was expanded with the addition of a real time distortion analyzer and virtual chart recorder.
SoundCheck’s new Sequence Editor revolutionizes the way that audio tests are written. For the first time, engineers, without programming knowledge are able to use a simple point-and-click interface to create advanced test sequences that can be run over and over again, saved, exported, modified and more. Although it has received a couple of major overhauls to increase functionality during its almost 25 year history, the sequence editor is still the cornerstone of SoundCheck’s test flexible development capabilities, and has facilitated the software’s expansion into multiple evolving markets.
Listen is first to incorporate an Autodelay algorithm. This automatically compensates for the delay between stimulus and response, displaying the result in time, samples and distance. It also measures signal polarity. Although the initial application of this algorithm was to measure the distance from and align the loudspeaker to the measurement microphone, these days it is invaluable for testing open-loop devices with huge latencies.
Normalized Distortion Measurement
Listen is the first to offer Normalized Distortion measurement. This normalizes distortion at the actual measured frequency to remove the influence of a non-flat frequency response on the percentage distortion calculation relative to the fundamental. This also makes it possible to measure distortion in an ordinary room or car without the room reflections affecting the measurement.
First Rub & Buzz Algorithm Using Higher Order Harmonics
SoundCheck is the first test system to detect Rub & Buzz by measuring over 200 high order harmonics both individually and simultaneously (rather than simple high pass filters which lump harmonics and noise together). This revolutionary method enables a high degree of selectivity in the analysis options, enabling specific faults to be identified and detected. For example, rub & buzz due to lead wires hitting a speaker cone usually occurs at the 50th harmonic and above, whereas a rubbing voice coil occurs mostly in 10-15th harmonic.
First SoundCheck System Sold
True to his word, the customer who had promised to buy a software-based audio measurement system if Steve could develop it, purchased the first-generation SoundCheck system. The photo shows SoundCheck version 1 in use at the customer site. Despite management changes and a company purchase, they are still using SoundCheck today and recently installed version 18!
HarmonicTrak algorithm developed
The HarmonicTrak algorithm is the first audio measurement algorithm to dramatically decrease measurement time by measuring the fundamental and the harmonics of a stepped sine stimulus in parallel (over 10x faster than a Brüel and Kjær 2012). We believe we are still the only people using this algorithm today.
Stweep™ algorithm developed
The Stweep™ algorithm uses fast stepped sine sweep measurements with an integer number of cycles at each frequency step. With frequencies changing at zero phase and amplitude, transition between frequencies is smooth, and the test signal sounds like a continuous sine chirp. This significantly reduces transducer settling time, resulting in faster and more accurate measurements. This is still one of the most widely used acoustic test signals in use today!
First Hardware-Agnostic, Soundcard-based Test System
Listen began the software-based audio test revolution when it became the first company to launch a commercial software-based audio test system that would work with any hardware and audio interface, eliminating the need for expensive proprietary hardware. The first version of SoundCheck used an off-the-shelf NI data acquisition card, and by mid 1996, Listen was the first company to offer a professional soundcard-based measurement system. Listen’s software has always worked with any soundcard/audio interface and any test hardware, keeping flexibility high and the cost of ownership low. Today, Listen offers its own audio interfaces with additional features to simplify audio test, as well as remaining compatible with, and supporting soundcards and data acquisition cards with ASIO, Core Audio, WASAPI, WDM, DAQmx and other communications protocols from multiple manufacturers.
Listen, Inc. Founded
Listen, Inc. is founded by Steve Temme, after a potential customer promised Steve that if he could build a software-based audio measurement system in 6 months that met his testing needs, he would buy it. He worked from the bedroom in his one-bedroom apartment until August 1998 when he moved the business to a colleague’s recording studio in Quincy, and a couple of years later to offices in Boston. These days, Listen operates from an 8,000+ square ft building in Boston’s South End.