- Since 1964 -
Your All Access Pass
gets attendees full access
to all content at AISE
at NO added fees
You need to be here!
Over 35 hours of quality content in two packed days means value, means ideas you can use, means a technical edge, means a business edge. It’s the edge YOU need to succeed as a company and as an individual professional.
Content is Always King at AISE!
Our content will be updated on a regualr basis. Check back often!
How to figure it out!
Content is broken down by category and color coded for easy identification in our Program and online.
Our full schedule will be posted in September 2017
Seminars, Papers, Forums and Advertorials
We want to create an event that does more than just look nice on a page. What are the creative companies of the future going to look like?
Preparing Speech Data for Testing of Automatic Speech Recognition Systems
by Jesse Gratke- Head acoustics
An essential component of testing Automatic Speech Recognition (ASR) systems is the creation of a suitable database of speech utterances and background noise scenarios. This seminar is designed to instruct attendees on how to properly record, playback, and process speech used in the testing of ASR systems. Seminar Outline: Introduction Speech Recording Methods Speech Playback Fundamentals Questions & Answers Dealing with Intellectual Property Copying, Piracy, and Theft Issues By J.Andrew McKinney There are treaties which extend Intellectual Property (“IP”) protection beyond US borders, and we will explore how the various forms of IP may extend across borders. Also, even a foreign based Company (like B), if selling in the US, is subject to US IP ownership and Tort claims, which means this alleged infringement should be examined from a transactional perspective (who, doing what, where and when). These issues and others will be addressed including comments and concerns of and by attendees. Because of the size and scope of the subject, there may be time constraints as to what issues we may address. Expect a lively and informative interactive discussion. This seminar will lead into the Forum "The Intellectual Property Dilemma" which will be immediately following this seminar.
The Intellectual Property Dilemma
ALMA Board of Directors
It has been a major issue for many years.
Various entities violate copyright law, or are accused of violating the law. Even when proven in a court of law, enforcement is at times more expensive than the lost revenue.
At times, enforcement is impossible regardless of time and money expended. What happens when a Member of ALMA accuses another Member of such behavior? How should your association deal with this issue? What would you do if confronted with the issue in your business? Please join ALMA International’s Board of Directors in a round table discussion about the many layers of consideration regarding this, and related issues. This will be an open and interactive forum. No names, or accusations will be permitted, but no other restrictions will be placed on the discussion. Look forward to a spirited debate as part of our annual Open Members Meeting on January 7, 2018 in NAPA A/B at 11AM. Green Speaker Design Reproduction with Maximum Efficiency By Wolfgang Klippel Besides horn-loaded compression loudspeakers, most direct radiating loudspeakers convert only a fraction of the electrical energy into sound. The low efficiency generates not only a heat problem but also shortens the battery lifetime in portable applications. There is a greater need to increase the output while reducing the electrical power consumptions. This workshop discusses the physical limitations and the practical design of passive transducers, where efficiency is one of the most important design goals. This approach requires adaptive control based on digital signal processing to cope with nonlinearities and other time varying properties inherent in efficient transducers. The amalgamation of passive transducer design, digital signal processing and system integration requires reliable loudspeaker models and meaningful characteristics in order to simplify the communication and to evaluate the product. This new paradigm is illustrated with woofers used in automotive and professional applications and with micro-speakers used in personal sound equipment. Learning from Production By Robert Werner In many cases, production testing is a necessary burden required to make sure that the audio product fulfills the specification sheet data and that no defective units are shipped to the customer. At the same time it is necessary to ensure that the yield rate is as close to 100% as possible to be profitable and save resources. Modern production test systems provide fast and reliable means for automatic quality control of audio systems at the end of the line or for incoming goods inspection, based on objective parameters. Limits are defined to make a simple Pass/Fail decision or to assign different quality grades. At this stage the task of production testing usually ends. The resulting data is mostly omitted or only stored for traceability purpose. Nowadays, where “big data” plays an important role even in our everyday life, this is a waste of opportunities although the full complexity of the term may sound intimidating. At the same time normal process control strategies are not always suitable for the specifics of loudspeakers. This workshop discusses how you can exploit the production test data of your audio product, both on and off the production line. It addresses topics like optimal limit definition based on long term statistics, selecting “golden units”, using smart algorithms (cluster analysis) for root cause analysis (“golden defects”) and automatic on-line classification, handling rare error patterns, reducing data complexity, closing the QC-R&D loop, control rules and trend analysis for early rejects detection and smart process control.
Testing of Automotive Audio Systems
by Jesse Gratke- Head acoustics
A method of a Jury evaluation of a vehicle audio system will be presented. The jury results provide insight on how the audio system was perceived and how it may or may not relate to the traditional objective metrics. Some research on tuning an audio system based on a method related to human equal loudness contours vs a method with a flat frequency response will also be presented.
The most typical ways of measuring the audio system in an automobile require traditional measurements that include: Frequency response (FR), Total Harmonic Distortion (THD), and sound pressure level (SPL). The FR, THD, and SPL are usually measured at nominal and then stepped up to maximum. However, most audio systems are typically tuned and validated subjectively. Therefore, a set list of tracks and music are listened to and recorded in the vehicle. These recordings can then be accurately played back using a high-fidelity playback system and a jury analysis can be conducted. The Future of End Of Line Testing by Jonathan Novik- Avermetrics There are smart-phones, smart-TVs, smart-everything, except smart-speakers. The basic raw speaker is still a “dumb” component that doesn’t communicate back to the systems they are used in. There is every reason to believe that this will not always be the case. Many of the technologies being developed for the Internet-of-Things could easily find their way into speakers. This session will explore the concept of “smart” speakers, the benefits they could offer and how they translate into end-of-line testing. Jonathan Novick has assembled a diverse panel representing many parts of the supply chain to discuss these possibilities. How Limiting Audio Reproduction Bandwidth to 20 kHz May Also Be Limiting Your Potential Market By Dan Foley High-resolution audio, currently defined as a minimum sampling rate of 96 kHz, is now the rage in the audio industry. From its inception, the majority of audio manufacturers designed their products with a bandwidth not exceeding 20 kHz. Electronics, in particular converter-based products, can extend bandwidth to beyond 100 kHz but the vast majority of loudspeakers start rolling off above 20 kHz. This bandwidth limitation makes sense if musical instruments have little-to-no acoustic energy above 20 kHz. But what if there is energy present when music is being created? To determine if this is the case, Dan Foley worked with students from Worcester Polytechnic Institute to conduct an experiment to determine whether or not common musical instruments generate ultrasonic energy at typical microphone recording positions in a studio environment. A trumpet, violin, conga drum, nylon string guitar, and a soprano vocalist were individually recorded using a typical studio microphone as well as two well-known measurement microphones with a flat response to well above 20 kHz. The experiment results will be presented showing that many of these instruments do produce ultrasonic energy with some as high as 80 kHz. How these findings may impact the loudspeaker industry will also be discussed in regards to emerging high-resolution audio formats. Objective Testing of High-End Audio Systems By Brian MacMillan When looking at the high-end audio market, there’s a wide variety of products ranging from well- designed top-notch audio devices with solid technical specifications, to gadgets that appear rather esoteric and whose effects cannot be evaluated objectively. The latter are not in the scope of this paper but only audio devices and systems that are designed based on solid audio-engineering work. This paper describes best practice examples of objective measurements on high-end audio devices. This is seen from two views: The first view describes the typical measurements on different devices along the signal chain, from the signal source, to the amplifier, over the cables to the loudspeakers and also the room in which the system is installed. The second view covers the different measurements at different life-cycle stages of the various devices, from R&D, to incoming QC, end of line QC, installation of the system as well as service and maintenance. End of line test concepts to achieve and maintain yield and quality in high volume loudspeaker production By Gregor Schmidle Keeping up consistent yield and quality in a high volume loudspeaker production environment requires a well designed manufacturing and test processes. There are many possible reasons for variations. Within a single production line, variations can be caused by fluctuations in the quality of subcomponent material, changes in the manufacturing and assembly processes, as well as variance in environmental influences. Commonly, several production lines of the same type are operated in parallel, introducing line-to-line deviations. In many cases, these production lines are spread across multiple facilities. This paper describes the aspects of the loudspeaker end of line test system in the manufacturing process. The first goal is to avoid or at least minimize variation by design where possible. The remaining deviations should be constantly monitored and analyzed. Challenges of IoT Smart Speaker Testing By Daniel Knighten Quantitatively measuring the audio characteristics of IoT (Internet of Things) smart speakers presents several novel challenges. We discuss overcoming the practical challenges of testing such devices and demonstrate how to measure frequency response, distortion, and other common audio characteristics. In order to make these measurements, several measurement techniques and algorithms are presented that allow us to move past the practical difficulties presented by this class of emerging audio device. We discuss test equipment requirements, selection of test signals and especially overcoming the challenges around injecting and extracting test signals from the device. Evaluation of Audio Test Methods and Measurements for End- of-Line Loudspeaker Quality Control By Steve Temme In order to minimize costly warranty repairs, loudspeaker OEMS impose tight specifications and a “total quality” requirement on their part suppliers. At the same time, they also require low prices. This makes it important for driver manufacturers and contract manufacturers to work with their OEM customers to define reasonable specifications and tolerances. They must understand both how the loudspeaker OEMS are testing as part of their incoming QC and also how to implement their own end-of-line measurements to ensure correlation between the two. Specifying and testing loudspeakers can be tricky since loudspeakers are inherently nonlinear, time-variant and effected by their working conditions & environment. This paper examines the loudspeaker characteristics that can be measured, and discusses common pitfalls and how to avoid them on a loudspeaker production line. Several different audio test methods and measurements for end-of- the-line speaker quality control are evaluated, and the most relevant ones identified. Speed, statistics, and full traceability are also discussed. Fast Measurement Under Simulated Free Field Conditions By Wolfgang Klippel Evaluating the loudspeaker performance requires a measurement of the sound pressure output in the far field of the source under free field condition. If the available test room does not fulfil this condition, it is common practice to generate a simulated free field response by separating the direct sound from the room reflection based on windowing and holographic processing. This paper presents a new technique that performs a filtering of the measured sound pressure signal with a complex compensation function prior to common time and frequency analysis. The influence of room, nearfield and positioning error is compensated in the measured frequency response and nonlinear distortion characteristics. Different methods are presented for generating the compensation function based on a reference response measured under anechoic conditions and a test response measured under in-situ conditions. Benefits and particularities are demonstrated by practical measurements using different kinds of test signals. Why Testing Voice Coil Position in Production Line and How? By Robert Werner The performance of loudspeakers at high levels is limited by nonlinear and thermal mechanisms related to motor and suspension as well as geometric constraints. Such transducer characteristics reduce the acoustical output, affect the speaker system alignment, cause unstable behavior and create audible nonlinear distortions such as intermodulation and harmonic distortion. In the R&D process, it is important to understand the whole complexity and effect of nonlinear transducer characteristics for optimal design with respect to the target application and quality. In production this matter is less complex, but still crucial. Most parameters can be considered as consistent because they are defined by transducer design (e.g. motor geometry). However, basic characteristics such as voice coil position and suspension properties may vary significantly during production and as a consequence, seriously degrade the large signal performance of the delivered product. Testing nonlinear symptoms, such as total harmonic distortion and applying simple pass/fail limits may prevent shipping bad units, but it is not sufficient for diagnostics and early defect detection. For example, it does not indicate whether the distortion is caused by the motor or the suspension and how it can be fixed (direction and quantity of coil offset). This paper discusses which large signal transducer parameters are important in production, how they are measured and how they can be used for root cause analysis and process control in order to maximize yield. Interfacing large signal parameter testing in R&D and QC is addressed as well.