About
90
Publications
47,257
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
707
Citations
Citations since 2017
Introduction
My research focus is on two main things
- customization of information systems, especially in SaaS scenarios and the industrial internet of things
- application of deep learning techniques to computer vision, speech recognition and natural language understanding
We are conducting design science research by creating software, introducing it in companies and empirically investigating the effects.
Our current projects are
- Supporting team collaboration, esp. document creation in geografically distributed teams with an integrated intranet solution (AMiProSI)
- Creating a highly customizable manufacturing execution system (HiCuMES) and extending it with Smart Production Planning and Scheduling (SPPS) capabilities
- Evaluating open source deep neural networks Mozilla DeepSpeech and BERT
Additional affiliations
March 2011 - present
October 2010 - present
September 2010 - present
iisys - Institute for Information Systems
Position
- Head of research group
Description
- Open Source, System Integration, Enterprise Knowledge Infrastructure, Unified Information Access, Building Automation esp. Smart Heating
Education
October 2002 - April 2006
May 1996 - April 2000
Publications
Publications (90)
We present ASR Bundestag, a dataset for automatic speech recognition in German, consisting of 610 hours of aligned audio-transcript pairs for supervised training as well as 1,038 hours of unlabeled audio snippets for self-supervised learning, based on raw audio data and transcriptions from plenary sessions and committee meetings of the German parli...
During the last decade, a lot of procedure models for the introduction of industry 4.0 (I40) or industrial internet of things (IIoT) solutions have been proposed, especially in the German literature. Since they target all kinds of I40 projects they necessarily must be somewhat generic and are missing important details especially for the “implementa...
Autoregressive TTS models are still widely used. Due to their stochastic nature, the output may vary from very good to completely unusable from one inference to another. In this publication, we propose to use the percentage of completely correct transcribed sentences (PCTS) of an ASR system as a new objective quality measure for TTS inferences. PCT...
The amount of freely available systems for automatic speech recognition (ASR) based on neural networks is growing steadily, with equally increasingly reliable predictions. However, the evaluation of trained models is typically exclusively based on statistical metrics such as WER or CER, which do not provide any insight into the nature or impact of...
Autoregressive TTS models are still widely used. Due to their stochastic nature, the output may vary from very good to completely unusable from one inference to another. In this publication, we propose to use the percentage of completely correct transcribed sentences (PCTS) of an ASR system as a new objective quality measure for TTS inferences. PCT...
While many speech synthesis systems based on deep
neural networks are thoroughly evaluated and released for free
use in English, models for languages with far less active speakers
like German are scarcely trained and most often not published
for common use. This work covers specific challenges in training
text to speech models for the German l...
The increasing availability of audio data on the internet leads to a multitude of datasets for development and training of text to speech applications, based on deep neural networks. Highly differing quality of voice, low sampling rates, lack of text normalization and disadvantageous alignment of audio samples to corresponding transcript sentences...
Software integration, especially in Open Source Software (OSS), suffers from technical debt as software evolves over time. Evolvement often ends in incompatibility issues like changed or removed APIs. The problems increase, as more components need to be integrated. This article analyzes common problems in OSS integration with respect to technical d...
The increasing availability of audio data on the internet lead to a multitude of datasets for development and training of text to speech applications, based on neural networks. Highly differing quality of voice, low sampling rates, lack of text normalization and disadvantageous alignment of audio samples to corresponding transcript sentences still...
Reading text aloud is an important feature for modern computer applications. It not only facilitates access to information for visually impaired people, but is also a pleasant convenience for non-impaired users. In this article, the state of the art of speech synthesis is presented separately for mel-spectrogram generation and vocoders. It conclude...
Virtual reality (VR) is on the edge of getting a mainstream platform for gaming, education, and product design. The feeling of being present in the virtual world is influenced by many factors and even more intriguing a single negative influence can destroy the illusion that was created with a lot of effort by other measures. Therefore, it is crucia...
Deep learning is moving more and more from the cloud towards the edge. Therefore, embedded devices are needed that are reasonably cheap, energy-efficient and fast enough. In this paper we evaluate the performance and energy consumption of popular, off-the-shelf commercial devices for deep learning inferencing. We compare the Intel Neural Compute St...
Deep learning has revived the field of automatic speech recognition (ASR) in the last ten years and pushed recognition rates into regions on par with humans. Applications like Siri, Amazon Alexa and Google Assistant are very popular, but have inherent privacy problems. In this paper, we evaluate state of the art open source ASR models regarding the...
Deep learning is moving more and more from the cloud towards the edge. Therefore, embedded devices are needed that are reasonably cheap, energy-efficient and fast enough. In this paper we evaluate the performance and energy consumption of popular, off-the-shelf commercial devices for deep learning inferencing. We compare the Intel Neural Compute St...
Sprachsteuerung wurde mit Siri populär und ist dank Alexa dabei unseren Alltag weiter zu vereinfachen. Sprachassistenten und Chatbots können aber auch in Unternehmen selbst, oder in der Kommunikation mit deren Kunden gewinnbringend eingesetzt werden. Dank künstlicher Intelligenz wird die Qualität solcher Lösungen immer besser. Man sollte jedoch nic...
Virtual reality (VR) is on the edge of getting a mainstream platform for gaming, education and product design. The feeling of being present in the virtual world is influenced by many factors and even more intriguing a single negative influence can destroy the illusion that was created with a lot of effort by other measures. Therefore, it is crucial...
We created a Web-based software to support the main production steps from getting the production order, setting machine parameters, checking product quality until confirming the number of produced goods and wastage. Production workers got one tablet per machine to interact with the system.
Manufacturing execution systems (MES) are the central integration point for the shop floor. They collect information about customer orders from ERP systems (enterprise resource planning), calculate the best plans for production orders and their assignment to machines and monitor the execution of these plans. It is therefore also the central data hu...
Manufacturing execution systems (MES) are the central integration point for the shop floor. They collect information about customer orders from ERP systems (enterprise resource planning), calculate the best plans for production orders and their assignment to machines and monitor the execution of these plans. It is therefore also the central data hu...
IT support for collaboration in knowledge-intensive processes has gone through quite some change since the beginning of the century and is providing more and more support for knowledge workers. Although single systems are getting easier to use, they are often not replacing former systems but accompany them, which makes the overall system landscape...
Das Privatleben in der digitalen Welt: Wie wird es sich verändern? Welche Produkte werden wir nutzen? Was können sie heute schon, welche Funktionen und Fähigkeiten werden sie in naher Zukunft haben? Wird unser Leben dadurch angenehmer oder sollten wir uns vor so einer digitalen Zukunft fürchten? Anhand vieler Anwendungsbeispiele werden sie verstehe...
The capability to operate cloud-native applications can generate enormous business growth and value. But enterprise architects should be aware that cloud-native applications are vulnerable to vendor lock-in. We investigated cloud-native application design principles, public cloud service providers, and industrial cloud standards. All results indica...
Purpose – Technologies such as smart advisors, smart factories and the “Internet of things” have again sparked large interest in IT-enabled knowledge management. Although there have been numerous case studies and empirical studies on how the management of knowledge in large organizations has benefitted from these technologies, we have only scarce i...
One could think that knowledge management (KM) is already fully digitized and therefore the discussion about digital transformation is already too late here. Documents are mostly created digitally for quite some years. Communication is happening mostly in digital form and also cooperation is often fostered by electronic media no matter whether it’s...
The capability to operate cloud-native applications can create enormous business growth and value. But enterprise architects should be aware that cloud-native applications are vulnerable to vendor lock-in. We investigated cloud-native application design principles, public cloud service providers, and industrial cloud standards. All results indicate...
Docker provides a good basis to run composite applications in the cloud, especially if those are not cloud-aware, or cloud-native. However, Docker concentrates on managing containers on one host, but SaaS provi¬ders need a container management solution for multiple hosts. Therefore, a number of tools emerged that claim to solve the problem. This pa...
Über die Enterprise-Tauglichkeit freier Software lässt sich je nach potenziellem Einsatzgebiet
sicherlich trefflich streiten. Schaut man sich die verfügbaren Ansätze jedoch genauer an, sieht es
für Open-Source-Software gar nicht so schlecht aus.
Der Hype um das Container-Werkzeug Docker ist auch drei Jahre nach seiner ersten Veröffentlichung ungebrochen. Docker adressiert sowohl Entwickler, die einen einfachen Weg suchen, um ihre Anwendungen zu testen und anschließend in die Produktivumgebung zu bringen, als auch Cloud-Betreiber, die durch die Verwendung von Containern statt Virtuellen Mas...
The Semantic Web has matured from a vision and research area of a few AI specialists to an important technology being used in a large number of research and a few practice projects. Most building blocks of the Semantic Web stack are filled with concrete technologies and W3C standards, but there are still enough areas for research. However, even wit...
In the present mobile climate, both companies as well as their consumers are able to execute various
processes at any time in any location using a myriad of wireless technologies. Organizational trends
demonstrate that mobile devices are being utilized exponentially more to solve problems from the
strategic planning level to the operational level t...
NoSQL and especially graph databases are constantly gaining popularity among developers as they promise to deliver superior perfor-mance when handling highly interconnected data compared to relational databases. Apache Shindig is the reference implementation for OpenSocial with a highly interconnected data model. However, it had a relational databa...
The paper presents results from an empirical study about the use of a single sign-on (SSO) system in an integrated open source system landscape for supporting team collaboration. A portal solution, enterprise content man-agement system, groupware, business process management and enterprise search engine are used. The investigation shows that althou...
Usually, knowledge workers are said to not benefit from business process management (BPM) systems, since their main tasks are weakly structured and not representable by a workflow. However, not all of their tasks are equally weak structured, and with adaptive case management (ACM) solutions, a new category of tools came up to support those processe...
This paper presents a concept for integrating social media functionality into existing collaboration systems and using this functionality to also integrate the collaboration systems themselves. Social media functionality is used in the areas contents, contacts and communication to support knowledge sharing and knowledge maturing. Using a micro-serv...
After the human-oriented, technology-oriented (documents), process-oriented and social KM phases, KM support now needs integration of those beneficial technologies instead of hyping one and neglecting the other. This paper proposes a highly modular yet integrated approach to support KM instruments and shows how this approach is implemented based on...
EN: In 2015, two conferences are brought together under the umbrella of „knowledge communities“ with complementary thematic priorities. While the GeNeMe deals with organisational and technical perspectives in the context of Virtual Enterprises, Communities & Social Networks, the ProWM provides a broad and inclusive overview of the organisational, c...
Docker provides a good basis to run composite applications in the cloud, especially if those are not cloud-
aware, or cloud-native. However, Docker concentrates on managing containers on one host, but SaaS provi-
ders need a container management solution for multiple hosts. Therefore, a number of tools emerged that
claim to solve the problem. This...
In this paper, we report about benchmark experiments and results from optimizing database connectivity for querying social networking data from Apache Shindig in a Neo4j database. We built on our experiments from [1] and tried to improve performance of the current RESTful http connection in comparison to JDBC in order to fully utilize performance b...
Knowledge management represents a key issue for both information systems’ academics and practitioners, including those who have become disillusioned by actual results that fail to deliver on exaggerated promises and idealistic visions. Social software, a tremendous global success story, has prompted similarly high expectations regarding the ways in...
Wissensmanagement stellt ein Schlüsselelement sowohl für Wissenschaftler als auch für Praktiker in der Wirtschaftsinformatik dar. Das gilt auch für jene, die von den bisher erzielten Resultaten, die den übertriebenen Versprechungen und idealisierten Visionen nicht gerecht werden konnten, desillusioniert sind. Social Software hat als globale Erfolgs...
Web applications and their architecture have changed significantly in the last decade and nowadays utilize asynchronous techniques such as Ajax or Comet to enable real time data communication and dynamic rendering. They further use HTML5 features such as drag&drop, canvas, Web workers and application cache in order to behave like desktop applicatio...
In this paper we describe our experiences with setting up a single signon enabled intranet with open source software and then making it accessible over the internet using a reverse proxy. During this process, we encounter several issues. We describe those, discuss possible solutions and present our final setup.
NoSQL and especially graph databases are constantly gaining popularity among developers of Web 2.0 applications as they promise to deliver superior performance when handling highly interconnected data compared to traditional relational databases. Apache Shindig is the reference implementation for OpenSocial with its highly interconnected data model...
Knowledge management is more and more happening in social environments, supported by social software. This directly changes the way knowledge workers interact and the way information and communication technology is used. Recent studies, striving to provide a more appropriate support for knowledge work, face challenges when eliciting knowledge from...
KM tools, especially ICT-based ones are often introduced in companies without having analyzed the
problem and requirements in detail, paying respect to accompanying organizational measures or even
having a proper KM strategy. In this book chapter, a framework for analyzing KM requirements and
choosing useful tools accordingly is presented which...
Knowledge management (KM) instruments increasingly utilize social software applications which provide an opportunity to connect human and technology orientation to improve knowledge activities. Social systems, such as wikis, micro-blogs or organizational social networks, support several knowledge purposes, e.g., to create and develop knowledge. Eve...
After the hype around Web 2.0, organizations have revisited their efforts on knowledge management in order to create beneficial effects from their employees’ activities in social environments. While there is not much doubt about the positive effects on crea-tivity and flexibility, handling the variety of knowledge work environments that are in use...
Knowledge Management [31] is a social activity. More and more organizations use social software as a tool to bridge the gap between technology- and human-oriented KM. In order to create interoperable, transferable solutions, it is necessary to utilize standards. In this paper, we analyze which standards can be applied and which gaps currently exist...
Access to structured information in DBMS which is usually done using Business Intelligence tools and access to unstructured information in document and content management systems is going to be unified. First steps are already done and more to come soon.
Even though there is abundant literature on successful cases of organizations applying knowledge management (KM) measures, many KM initiatives have failed to achieve their knowledge and business goals. In order to foster decisions about the design of such initiatives, information is required on success factors and barriers when selecting KM measure...
This paper is proposing a list of knowledge management instruments based on a thorough literature review considering English and German sources, a strict definition for km instruments and a new categorization scheme that allows identifying potential duplicates or missing pieces.
Der vorliegende Beitrag berichtet von den Erfahrungen des Autors in der Verwendung des Wissensreifungsmodells bei der technisch geprägten Wissensmanagement-Beratung, um Vorhaben zu strukturieren, Schwerpunkte zu setzen, Lücken in der technischen Unterstützung der Wissensarbeit zu identifizieren und technische Unterstützung für einzelne Reifungsphas...
Auch wenn er nicht mehr die Bezeichnung Office im Namen trägt, so rückt Microsofts SharePoint Server doch weiter an die Office-Clients heran. Während sich der Hersteller immer offensiver um die – auch kostenlose – Verbreitung seiner Client-Programme bemüht, muss sich SharePoint mehr und mehr als Umsatz- und Hoffnungsträger der Redmonder Office-Spar...
Information of all kinds, tasks, events, emails and documents, should be treated uniformly, no matter whether they are stored in the intranet or on the desktop of an information worker. In DMS and CMS it is quite common these days that information are annotated with metadata, whereas on desktop PCs, meta data support is only available in isolated a...
Even though there is abundant literature on successful cases of organizations applying knowledge management (KM)
instruments, many KM initiatives have failed to achieve their knowledge and business goals. In order to foster decisions
about the design of such initiatives, information is required on success factors and barriers when selecting KM inst...
Success of an organization is increasingly dependent on its capability to create an environment to improve the productivity of knowledge work. This book focuses on the concepts, models and technologies that are used to design and implement such an environment. It develops the vision of a modular, yet highly integrated enterprise knowledge infrastru...
Chapter 2 gave an overview of the basic information and communication infrastructure which can be found in many organizations.
Advanced services to support knowledge work are installed on the basis of this infrastructure. However, before we can touch
specific knowledge services that directly target knowledge work processes, we have to first discuss...
The knowledge services discussed in the previous chapter should not be Overview isolated from other applications a knowledge worker uses. On the one hand, users want access to them from within their usual work environment and they want access any time and any place. Therefore, mobile access and desktop-integrated access play an important role for u...
Infrastructure is a term used for a variety of things. The most common Overview contexts for the term in everyday life are roads, water and wastewater, electricity and telecommunication. The basis for an EKI is the network infrastructure. This starts with laying cables into the floors and/or the walls of organization’s buildings or setting up base...
In chapters one to five, we presented the vision of an EKI and reviewed existing technologies that can be used to realize parts of this vision today. In this chapter, we will take up the introductory thoughts on knowledge work, knowledge management, KM instruments, IS architecture and EKIs and review results from both theoretic and empirical resear...
Integration services as presented in chapter 3, 151ff represent the funda- Overview ment that draws together contents from varying sources and offers a semantically and functionally, more or less cleansed and validated knowledge basis on which the core services can be built that differentiate an EKI from a basic information infrastructure. These se...
Information and communication technologies (ICTs) to support handling Knowledge of knowledge in organizations have been discussed for quite a long time. In the beginning of applying ICTs, it was routine work that was first prone to automation. ICTs gradually broadened their scope and have been extended to embrace highly valued work that years befor...
Success of an organization is increasingly dependent on its capability to create an environment in order to improve productivity of knowledge work. This book focuses on the concepts, models and technologies that are used to design and implement such an environment. It develops the vision of a modular, yet highly integrated enterprise knowledge infr...
The growing interest in location-based services and geo data gives organizations the chance to connect them to their business processes in order to increase efficiency and decision quality of knowledge workers. However, the potential of geo data and technologies in organizational settings is seldomly used yet. This is especially true for knowledge...
To study the interfaces between business processes and knowledge processes like knowledge sharing and understand influencing factors it is important to have a good model of both processes. The goal of this paper is to develop a model of knowledge sharing that is based on a process definition of knowledge sharing and identifies context areas for inf...
Organizations these days have a wealth of electronic resources scattered
across a number of systems. These resources are implicitly linked to each other by having the same authors, referencing the same processes or projects or discussing the same topics. To support users in retrieving related documents together to enhance the effectiveness of knowl...
Technical article addressing application areas of XSLT and XSL-FO in document management
Technical article showing that implementation of standards like XML-based office file formats is always an interpretation of the specification and leads to incompatible details when exchanging documents between different applications.
Second draft of my PhD thesis.
Published in: Ruhland, J., Wachsmuth, T. (eds.): Viertes Interuniversitäres Doktorandenseminar Wirtschaftsinformatik, Jenaer Schriften zur Wirtschaftswissenschaft 26/2003, Jena, Dezember 2003
Presentation of the first draft of my PhD thesis proposing the suitability of multi agent-based simulation as a means to model knowledge work in software development companies.
Showing how different services add up to deliver a satisfying customer experience in B2B market places.
Projects
Projects (5)
Transferring the idea of "mass customization" (customer-specific individual production with methods of efficient mass production) to the introduction of IT systems.
Support for manufacturing SMEs in the digital transformation.
Expansion of company networks into innovation hubs where not only loose experiences are exchanged but concrete ideas for the digital transformation of the members are systematically developed.
Closer integration of universities and companies in order to use existing knowledge to strengthen the economy.
The project is about technology transfer to small and medium-sized businesses with a special focus on machine learning. The project is a frame for several smaller transfer activities that usually run 6-12 months and involve the research institute, an application partner (often manufacturing industry) and an IT company.
To learn about the practical applicability of our activity mining algorithm in order to make implicit knowledge about document creation processes explicit. To do so, we create an integrated intranet solution to support team collaboration and store activities of people with the different systems. They serve as a basis of our evaluation.